THE AABB BLOOD SURVEY REPORT Final

   The 2013 AABB Blood Collection, Utilization, and Patient Blood Management Survey Report

Report Authors Barbee I. Whitaker, PhD Srijana Rajbhandary, BDS, MPH Andrea Harris, MPH

Citation

 AABB, December 18, 2015

   Table of Contents

List of Tables...... v

List of Figures ...... vii

1. Executive Summary...... 1 US Blood Collection ...... 1 US Blood Utilization ...... 2 Patient Blood Management (PBM) ...... 3

2. Key Findings ...... 5

3. International Report ...... 9 Blood Collected and Produced ...... 9 Blood Transfusion ...... 10

4. Blood Collected and Produced in AABB Member Facilities:  Trends in US Collections ...... 11 Total WB/RBC Collections...... 11 Whole Blood (WB) ...... 11 Red Blood Cell (RBC) Apheresis...... 12 RBC Distributions ...... 14 Non-RBC Components ...... 14 Platelets ...... 14 Plasma...... 15 Cryoprecipitate ...... 17 Granulocytes ...... 18 Blood Center Inventory ...... 18

5. Blood Transfused by US-Based AABB Member Hospitals ...... 21 WB and RBCs Transfused ...... 21 Pediatric Transfusions ...... 21 Transfusion Recipients...... 22 Non-RBC Components Transfused ...... 22 Outdated Units ...... 26

iii THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

6. Patient Blood Management (PBM) in AABB Member Facilities ...... 29 PBM Interventions ...... 31 Measure of Intervention Success ...... 34 Informed Consent ...... 35 Type and Screen before Procedures ...... 36 Transfusion Thresholds ...... 36

7. Current Issues in Blood Collection and Donor Screening ...... 43 Donor Hemovigilance ...... 45

8. Current Issues in Blood Transfusion ...... 47 US Collection Trends ...... 47 Blood Availability ...... 48 Blood Use...... 50 Crossmatch Procedures ...... 52 Red Cell Age...... 52 Platelet Age...... 52 Hemovigilance ...... 52

9. Component Costs ...... 57 RBCs ...... 57 Plasma ...... 57 Apheresis Platelets ...... 58 Reimbursement ...... 58 Summary ...... 61

10. Component Modification ...... 63 Leukocyte Reduction ...... 63 AABB Member Transfusion of Irradiated and LR Components...... 63

11. Appendix I: Methods ...... 69 Survey Population...... 69 Non-US AABB Member Response...... 69 US Military Blood Program Response ...... 69 AABB US Member Data Analysis ...... 69 Response Rates and Weights...... 69 Imputation of Data Items...... 72 Characterization of Respondents...... 72 Comparison with Previous Surveys ...... 73 Data Quality Control ...... 74 Statistical Analysis...... 74 Limitations ...... 74

12. Appendix II: Glossary of Terms ...... 75

13. Acknowledgements ...... 77

iv THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

   List of Tables

Table 4-1. Estimated 2013 Collection and Transfusion by AABB US Member  Blood Centers and Hospitals for Whole Blood (WB) and Red Blood  Cells (RBCs) (expressed in thousands of units) ...... 13 Table 4-2. Estimated 2013 Collection and Transfusion by AABB US Member  Blood Centers and Hospitals for Non-RBC Components (expressed  in thousands of units) ...... 16 Table 5-1. Pediatric Transfusions by US Blood Centers and Hospitals in 2013  (expressed in thousands of units) ...... 22 Table 5-2. Outdated Components as a Percentage of the Total Number of Units  of Each Component Distributed for Transfusion in 2013 ...... 27 Table 6-1. Reporting Hospitals Policy for Platelet Threshold for Transfusion ...... 38 Table 7-1. Donor Adverse Reaction Rate by AABB Facility and Procedure Type. . . . . 45 Table 8-1. Transfusion-Related Adverse Reactions Reported to the Transfusion  Service ...... 54 Table 9-1. Mean AABB Member Hospital Amount ($) Paid per Selected  Component Unit in 2011-2013 ...... 58 Table 9-2. Average AABB Member Hospital Component Cost ($) by Surgical  Volume, 2013 ...... 59 Table 9-3. CMS Hospital Outpatient Prospective Payment System Rates for  Selected Blood Components ...... 60 Table 10-1. Blood Components Modified to Achieve Prestorage Leukocyte  Reduction in US AABB Facilities ...... 64 Table 10-2. Change in Number of Blood Components Modified to Achieve  Pre-Storage Leukocyte Reduction by Facility Type from 2011 to 2013 . . . 66 Table 10-3. Estimated Number of Blood Component Units Modified by Irradiation  or Leukocyte Reduction and Transfused by US AABB Member  Facilities in 2013 ...... 67

v THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

Table 10-4. Total Number of Irradiated and Leukocyte Reduced Red Blood Cell  (RBC) Units Transfused by US AABB Member Hospitals in 2013,  compared with RBC Units Transfused by AABB Member Hospitals  in 2011 ...... 68 Table A-1. AABB Blood Survey 2013: Survey Population and Response Rates— Non-US Members ...... 70 Table A-2. AABB Blood Survey 2013: Survey Population and Response Rates— US Military Blood Program ...... 70 Table A-3. AABB Blood Survey 2013: Survey Population and Response Rates ...... 70 Table A-4. Response Rate Summary and Comparison with Previous Surveys ...... 71 Table A-5. Weights for Blood Centers ...... 71 Table A-6. Weights for 2013 Hospitals...... 72 Table A-7. Calibrated Average Final Weight by USPHS Region and Inpatient  Surgical Volume – 2011 AABB Hospital Respondents...... 73

vi THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

   List of Figures

Figure 1-1. Growth in use of RBC apheresis technology ...... 2 Figure 4-1. Allogeneic and autologous WB and RBC collections, 1989-2013 ...... 12 Figure 4-2. Apheresis procedures by primary component ...... 15 Figure 4-3. Apheresis Platelets as Percent of Total Platelets Produced ...... 17 Figure 4-4. Sources of plasma for transfusion...... 18 Figure 4-5. Days of Group O Supply 2011 and 2013...... 19 Figure 5-1. Trends in platelet transfusion, 1997-2013 ...... 23 Figure 5-2. Types of plasma transfused in 2011-2013 ...... 24 Figure 5-3. Percent of plasma transfused as thawed plasma (2013 average 53%). . . . . 25 Figure 5-4. Percentage of Group O Allogeneic RBC Outdates ...... 28 Figure 6-1. Use of national transfusion guidelines, 2011-2013 ...... 31 Figure 6-2. Preoperative PBM interventions implemented, 2013 (n=199) ...... 32 Figure 6-3. Intra-operative PBM interventions implemented, 2013 (n=226) ...... 33 Figure 6-4. Postoperative PBM interventions implemented, 2013 (n=203)...... 34 Figure 6-5. Methods hospitals use to measure the success of implementations  to improve PBM...... 35 Figure 6-6. Reporting hospitals policy for a hemoglobin threshold for transfusion  for hemodynamically stable non-bleeding patients ...... 37 Figure 6-7. INR threshold for transfusion of plasma ...... 39 Figure 6-8. Reported PTT threshold for bleeding patients (n=174) ...... 40 Figure 6-9. Use of prothrombin complex concentrates (4-factor or 3-factor) by  hospitals reporting a policy for their use ...... 40 Figure 7-1. Donor deferrals 2013...... 44 Figure 8-1. Trends in estimated rates of WB/RBC collection United States,  1980-2013...... 48 Figure 8-2. Trends in estimated rates of WB/RBC transfusion in the United States,  1980-2013...... 49 Figure 8-3. US allogeneic WB/RBC collections and transfusions, 1989-2013...... 49 Figure 8-4. RBC use by hospital service in 2013 ...... 51 Figure 8-5. Platelet use by hospital service in 2013 ...... 51 Figure 8-6. Adverse reactions by type, 2011 and 2013 ...... 55 Figure 10-1. US Transfusion of modified RBC components ...... 65

vii

THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

1. Executive Summary

The 2013 AABB Blood Col- vices, both in the United Due to the high response lection, Utilization, and States and internationally. rate of blood collectors, Patient Blood Management There was no sampling—all estimates and comparisons Survey (Blood Survey) was responding AABB institu- to previous data can be sponsored by AABB and tional members were made for the entire US conducted with AABB included, regardless of size blood collection system. member participation. The and location. However, as AABB institu- objectives of this new sur- tional members administer vey were to gather blood The overall response rate only a portion of the trans- collection and transfusion for the 2013 AABB Blood fusions in the United States, information about AABB Survey was 54.5% (625/ comparisons of blood trans- members, both within the 1147). The blood center fusions are limited to activ- United States and interna- response rate was 92.4% ity at AABB member tionally, make comparisons (73/79) and there was an hospitals reporting in 2011 with previous US surveys additional 50.0% response versus 2013. where possible, and pro- from member US military vide patient blood manage- collectors (2/4). The hospi- The international response ment (PBM) benchmarks for tal response rate was 51.7% was less robust, with 35.3% AABB members complet- (552/1068).* Response rates (6/17) of non-US blood ing the PBM section. The were comparable to the centers and 22.9% (8/35) of 2013 Blood Survey find- 2009 National Blood Col- non-US hospitals reporting. ings are reported here in lection and Utilization Sur- It is possible that the low total, replacing Part I of the vey (NBCUS)† and slightly response was due to inade- 2013 Blood Survey Report higher than responses by quate communication to all released earlier and includ- AABB member institutions members that the survey ing all results. Survey results for the 2011 NBCUS. scope included non-US from AABB members out- AABB members for the sur- side of the United States are vey year 2013. included. *Military hospitals were included in this group as they serve military and non-military populations. The facilities surveyed †Report of the US Department of US Blood Collection included all AABB member Health and Human Services. The blood collectors (both 2011 national blood collection and utilization survey report. The 2013 AABB Blood Sur- blood center and hospital- Washington, DC: US Department vey estimates that 13.6 mil- based facilities), hospital of Health and Human Services, lion whole blood (WB) and transfusion services, and Office of the Assistant Secretary red blood cell (RBC) units centralized transfusion ser- of Health, 2012.

Executive Summary 1 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT were collected, a significant to 13.8% of all WB/RBC fewer than reported in 2011 decline of 12.1% collections (Figure 1-1). (p=0.036). AABB member (p<0.0001) from 2011. In facility transfusion of 2013, blood centers were There was a statistically sig- plasma decreased signifi- responsible for the collec- nificant 27.1% increase in cantly by 9.9% (p=0.036), tion of 13.1 million units, or the distribution of WB- continuing the decline 96.1% of the supply; the derived (WBD) platelet reported in the most recent remaining 3.9% were col- concentrate units (164,000 national survey. lected by AABB hospitals. apheresis equivalent units) Therapeutic collections that individually and in pools. There were 15.4% more were not intended for trans- However, total platelet dis- platelets transfused by fusion have not been tributions decreased 2.4% AABB institutional mem- included in this survey. overall due to a 4.2% drop bers in 2013, totaling 1.3 in apheresis platelet distri- million platelet transfu- There was a 5.5% decrease butions. sions (p=0.0423). Increases in the total number of RBC in both apheresis platelet apheresis units collected, transfusions (12.2%) and including both allogeneic US Blood Utilization WBD platelet transfusions and autologous collec- (30.7%) contributed to the tions, to approximately 1.9 difference. There was a The estimated number of million units (p=0.001). The marked increase in the WB/RBCs transfused by proportion of blood col- number of cryoprecipitate AABB member hospitals lected in this manner, how- units transfused (66.2%; was 6.1 million units, 7.3% ever, continued to increase p<0.0001).

100%

90%

80%

70%

60%

50%

40%

30%

20%

10%

0% 1999 2001 2004 2006 2008 2011 2013

Traditional Allogeneic Collections RBC Apheresis Collections

Figure 1-1. Growth in use of RBC apheresis technology.

2 Executive Summary THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

Some reasons for these various transfusion policies fusion thresholds for other increases in WBD platelet and protocols. This is evi- patient populations and and cryoprecipitate transfu- denced by the survey components are discussed sion are the addition of crit- results reflecting that 23.7% in Chapter 6. ical care beds, acquisition of the reporting hospitals of other facilities, increases had implemented five or Of the AABB member facili- in liver and solid organ more pre-, intra-, or postop- ties responding to the ques- transplants, and the impact erative interventions associ- tion, 37.8% reported of massive transfusion pro- ated with PBM, and another having a PBM program, a tocols. 22.9% of hospitals reported slight increase when com- implementation of at least pared to 35.9% in 2011. three PBM interventions. Additional survey results Patient Blood expand on the baselines Management (PBM) Over 55.0% of reporting established with the 2011 AABB member hospitals NBCUS and have allowed The efforts of AABB to pro- indicated transfusion the development of individ- mote the adoption of PBM thresholds for RBC transfu- ual hospital benchmarks for initiatives are well reflected sion for the general inpa- 2013 data. in this year’s hospital tient population between 7 responses regarding their and 8 grams (g) of hemoglo- continued incorporation of bin per deciliter (dL). Trans-

Executive Summary 3

THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

2. Key Findings

The results of the 2013 million units in 2011 to 2011 (p<0.001). Non- AABB Blood Survey pro- 11.7 million units in US collectors reported vide information about 2013 (12.8% decrease). less than 1000 autolo- AABB member blood col- • Ten AABB members gous units collected, lection and transfusion ser- located outside of the suggesting that this vices and related activities United States reported practice is not common in the 2013 survey year. collections of 1.2 mil- outside of the United Participating members were lion units of WB/RBCs. States. predominantly from the • In the United States, • The number of units United States. Compari- 14.0 million individuals rejected for unaccept- sons of 2013 US member presented or registered able test results in the responses were made to the to donate in 2013, a United States decreased AABB member cohort from decline of 21.0% from by 8.2% to 90,000 in the 2011 NBCUS. Notable 2011. 2013 from 98,000 in findings from the 2013 • Of the people present- 2011 (p=0.008). How- AABB Blood Survey and ing to donate blood or ever, the proportion of comparisons with member blood products to US collections with these responses to the 2011 sur- blood collectors, 6.8 reactive infectious dis- vey are listed below. million allogeneic ease marker results was donors successfully 0.7%, the same reported gave blood compared in 2011 and 2008. Collection with 9.1 million alloge- • Units of WB/RBCs dis- neic donors in 2011, a carded for other rea- • Total WB/RBC collec- 24.9% decrease. sons by US blood tions by domestic AABB • The collection of RBCs collectors (not includ- members decreased sig- by apheresis in the ing outdated products) nificantly (p<0.0001) United States decreased was 629,000, a from 15.5 million units by 5.5% from 2.0 mil- decrease from 2011 of in 2011 to 13.6 million lion in 2011 to 1.9 mil- approximately 38.2% units in 2013 (12.1% lion in 2013 (p=0.001). (p<0.0001). decrease). • Autologous units col- • Allogeneic blood col- • Total WB Allogeneic lected in the United lection in the United collections (including States (manual and States among AABB directed) by domestic apheresis collections member blood centers AABB members combined) totaled and hospitals was  decreased significantly 55,000, a decrease of 65.5 units per 1000 per- (p<0.0001) from 13.4 47.1% compared with sons ages 16 to 64 in

Key Findings 5 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

2013 compared with reactions reported by WB/RBC transfusions of 76.2 units per 1000 per- blood collectors in 2013 59,000 units. sons the same ages 16 in the United States. • There were 13,000 to 64 in 2011. These occurred in autologous units of WB/ • The distribution of WB 0.13% of collection pro- RBCs transfused by US for transfusion as WB by cedures (unchanged AABB member hospi- domestic AABB mem- from 2011). Another tals in 2013, a decrease bers decreased signifi- 2700 severe reactions of 58.1% (p<0.001) cantly (p<0.0001) from were reported by non- compared with 2011. 48,000 units in 2011 to US blood collectors • Transfusion of aphere- 6000 units in 2013 (0.2% of reported col- sis platelets in US AABB (87.5% decrease). lection procedures). member hospitals • 5.2 million WB/RBC • In 2013, there were increased by 12.2% Group O units were dis- 176,000 more WBD from 1.0 million in tributed by AABB US platelet concentrates 2011 to 1.1 million member blood collec- distributed in the United units in 2013. tors in 2013, comprising States (either as individ- • Plasma transfusions in approximately 41.7% of ual units or combined US AABB member hos- all WB/RBC distribu- within pools); this was a pitals decreased 9.9% tions. In 2011, 37.6% of 27.4% increase from for a total of 1.8 million WB/RBC distributions 2011. units transfused. were Group O, a total of • There were 10,103,000 • Extrapolation of the US 5.6 million units. WB/RBC units col- AABB membership data • The number of aphere- lected, prepared, or to the total number of sis platelets distributed modified to achieve pre- transfusions performed in the United States in storage leukoreduction throughout the United 2013 stayed essentially in 2013 in the United States, yields a rate of the same, decreasing States, comprising approximately 40.3 WB/ from 2.1 to 2.0 million 78.5% of the total WB/ RBC units per 1000 (US units; however, this RBC products available population), an 8.4% 4.2% decrease was sta- for distribution. decrease from 44.0 tistically significant units per 1000 esti- (p=0.015). mated in 2011. • There were a total of 4.3 Transfusion • Approximately 37.8% of million units of plasma responding AABB mem- produced for transfusion • In 2013, the total WB/ ber hospitals reported in the United States in RBC units transfused by having a PBM program 2013, a decrease of US AABB member facil- in 2013. 26.0% from 2011 ities was 6.1 million • The total number of (p<0.0001). Of these, units, a 7.3% decrease components transfused 3.5 million were actu- from the units transfused by US-based AABB ally distributed; this was by US AABB member member hospitals in 22.4% fewer units than facilities in 2011. 2013 was 10,962,000, a in 2011 (p<0.0001). • Eight AABB member 1.1% increase over the • There were 17,600 hospitals outside of the 2011 AABB member severe donor adverse United States reported transfusions.

6 Key Findings THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

• The number of irradi- 2011, for a total of to US-based AABB ated RBC units trans- 817,000 units (13.3% of member hospital trans- fused by US-based all RBC units trans- fusion services at a rate AABB member hospi- fused). of 2.5 per 1000 compo- tals decreased by 16.8% • Adverse transfusion nents transfused in in 2013 compared to reactions were reported 2013.

Key Findings 7

THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

3. International Report

The international (non-US) Most collections were man- platelets. There were AABB institutional member ual WB collections, totaling 106,000 WB-derived plate- response to the 2013 AABB 1,240,000 units, 53.3% of lets distributed, 76.0% of Blood Survey was 35.3% which were collected at which were distributed in (6/17) for blood centers and mobile blood drive sites. pools. 22.9% (8/35) for hospitals. There were only 600 units Members reported from 10 of autologous WB collected countries. Weights were not (0.05% of collections), a A total of 466,000 units of applied and estimations of smaller proportion than US plasma were produced for representativeness were not autologous collections transfusion. This total made. (0.4%). There were also includes WB-derived fresh 10,000 allogeneic RBC frozen plasma (FFP; units collected by aphere- 112,000 units), plasma fro- Blood Collected and sis. Collections were zen within 24 hours after Produced reported from 641,000 allo- phlebotomy (137,000 geneic donors. units), cryoreduced plasma The respondents, four of (57,000 units), liquid which were hospital collec- There were 2700 severe plasma (89,000 units), and tors, reported collection of blood donor reactions plasma from apheresis col- 1,249,000 units of WB and reported, giving an aggre- lections (71,000 units). A apheresis RBCs, with blood gate rate of 0.2% reactions total of 50,000 plasmapher- centers collecting 1,216,000 per collection procedure. esis procedures generated units, or 97.4%, of the total. 60,000 units of apheresis The number of units There were 59,000 platelet- plasma for transfusion; rejected for unacceptable pheresis procedures other apheresis procedures test results was 8000 units. reported by non-US AABB produced 11,000 units of There were an additional members in 2013. These plasma. Of all types of 83,000 units discarded for procedures yielded 70,000 plasma, there were 22,000 other reasons than testing. apheresis platelet units, units of Group AB plasma The remaining 1,158,000 98.7% of which were col- distributed for transfusion. units were available for lected in plasma. Most col- transfusion—representing lectors reported a single In 2013, 87,000 individual 92.7% of the total units unit yield per apheresis pro- units of cryoprecipitate collected. cedure. Blood centers col- were produced and 411 lected 96.0% of apheresis granulocyte units prepared.

International Report 9 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

Blood Transfusion United States, of these, reported by non-US facili- 55.4% were apheresis ties. There were 18,000 In 2013, nine AABB mem- platelet transfusions. Only electronic crossmatch pro- ber institutions located out- one hospital reported trans- cedures, 17.8% of the total side the United States fusing both apheresis plate- procedures reported. Man- reported transfusion data. lets and WB-derived ual tube serologic proce- Transfusions of WB/RBCs of platelets—most hospitals dures accounted for 71.3% all donation types totaled transfused either WB- (72,000 procedures). Only 59,000 units. All of these derived platelets or aphere- 6.9% were reported to be institutions reported a pol- sis platelets. A total of automated serologic cross- icy of universal leukoreduc- 19,000 units of plasma, match procedures. The tion (LR, 66.7%) or of 3000 units of cryoprecipi- remaining procedures were transfusing LR components tate, and 98 units of granu- not categorized. To calcu- to particular types of locytes were reported late the crossmatch-to patients, including neo- transfused in 2013. transfusion ratio (C:T), the nates, oncology patients, total number of allogeneic and others (33.3%). Five hospitals reported for- WB/RBC units transfused mal, established programs (59,000) was used as the for patients who refuse any denominator. The overall Of the RBC units transfused, C:T ratio was 1.7 cross- 8000 units were Group O- or all blood components. Most facilities (7/9) reported match procedures per unit positive (13.6%) and 2000 transfused. units were O-negative having a Transfusion Safety (3.4%). There were fewer Officer (TSO) or Patient than 100 autologous trans- Blood Management Coordi- There were 280,000 patient fusions reported. There nator (PBMC) onsite. As the specimens reported submit- were 19,000 WB/RBC units reporting facilities repre- ted to blood banks for ABO reported as outdated. How- sented different countries, testing. There were 2000 ever, without a report of all their responses may not sample collection errors WB/RBC units collected indicate that TSOs and and 564 transfusion reac- and transfused by country, PBMCs are widespread tions reported. Of the trans- this number cannot be put globally; however, the con- fusion reactions, most were into context. cept of the TSO is modeled febrile reactions (44.5%) or on the staff role established mild allergic reactions in Canada. (39.2%); 7.1% were There were 9000 units of reported to be Transfusion platelets transfused by A total of 101,000 cross- Associated Circulatory AABB member hospitals Overload (TACO). located outside of the match procedures were

10 International Report THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

4. Blood Collected and Produced in AABB Member Facilities: Trends in US Collections

WB and RBC (by apheresis) previously had been The number of units collections for the survey reported separately, were rejected for unacceptable years 1989 through 2013 included in the total for test results decreased by are illustrated in Figure 4-1. allogeneic blood collection. 8.2% compared to 2011 Total collections reached a (p=0.008), proportionately high of 17.3 million units in less than the drop in collec- 2008.* For 2011 and 2013, Total WB/RBC tions. There were an addi- the figure shows total col- Collections tional 629,000 (±32,000) lections data restricted to units that were discarded AABB member facilities; for reasons other than test- these decreased signifi- The total WBD and aphere- ing, (eg, bag failures). Also, cantly (p<0.0001) between sis RBCs collected in the 12.9 million WB/RBC units 2011 and 2013 by 12.1% United States in 2013 was were reported available for to 13.6 million units. 13.6 million (±207,000) transfusion, 94.7% of the units (Table 4-1). Blood total units collected. centers collected 13.1 mil- Autologous collections con- lion units, or 96.1% of the tinued to decline from total. The remaining Whole Blood (WB) 2011. In 2013, autologous 527,000 units (3.9%), col- totals included 48,000 lected by hospitals, were a manual collections and slightly smaller proportion Donations of WB in 2013 7000 apheresis red cell col- compared to previous totaled 11.7 million lections, and amounted to years. Even RBC apheresis (±187,000) units. These less than 1.0% of total col- collections, which had pre- collections, reported lections. In 2013, collec- viously increased with according to the type of tions for the use of a every survey, declined sig- donation, are shown in designated patient, or nificantly by 5.5% in 2013. Table 4-1. Allogeneic dona- directed donations, which However, in 2013, RBC tions totaled 11.7 million, apheresis collections repre- of which 95.7% were col- *Report of the US Department of sented a larger fraction of lected by blood centers and Health and Human Services. The the total WB/RBC collec- 4.2% by AABB member 2009 national blood collection tions (13.7% compared hospitals. There was a and utilization survey report. with 12.7% in 2011). decrease in allogeneic Washington, DC: US Department donations between 2011 of Health and Human Services, and 2013 of 12.8% Office of the Assistant Secretary for Health, 2011. (p<0.0001).

Blood Collected and Produced in AABB Member Facilities: Trends in US Collections 11 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

18

16

14

12

10

Allogeneic 8 Autologous Millions of Units

6 AABB Allogeneic

AABB Autologous 4

2

0 1989 1992 1994 1997 1999 2001 2004 2006 2008 2011 2013

Figure 4-1. Allogeneic and autologous WB and RBC collections, 1989-2013.

Autologous, or self-directed, Red Blood Cell (RBC) While nearly all of the RBC WB units totaled 48,000 Apheresis apheresis collections were (±3000), a statistically for allogeneic use, a small significant decrease of number of units collected 52.0% compared to 2011 In addition to WB collec- by RBC apheresis were for (p<0.0001). Hospitals col- tions, 1.9 million (±49,000) autologous use (7000 lected 20.8% of all autolo- RBC units were collected units). gous units. by apheresis. Apheresis RBC collections in 2013 decreased by 5.5% in com- RBC apheresis collections There were only 6418 parison to 2011, when were performed largely in (±1000) units reported as there were 2 million RBC blood centers, accounting distributed as WB for trans- units collected by AABB for 98.9% of such units. In fusion in 2013 by AABB member facilities. There 2011, 116 blood centers member facilities, com- were 956,000 RBC aphere- and 38 hospitals reported pared to a significantly sis collection procedures, RBC apheresis collections. higher number in 2011 1.7% fewer than in 2011 In 2013, 72 blood centers (48,000 units; p<0.0001). (Figure 4-2). and 25 hospitals reported collecting RBCs by aphere- sis. The apparent decrease

12 Blood Collected and Produced in AABB Member Facilities: Trends in US Collections THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT <0.0001 10.036 87.5 <0.0001 % Change 2011-2013 p-value 2011 2011 Total % of Total Total of % Collections/ Transfusions Transfusions ood Centers and Hospitals for Whole Blood 6,118 311 99.8 6,584 –7. 2013 Total* ±95% CI Combined Blood Centers Hospitals ction and Transfusion by AABB US Member Bl WB Autologous (manual)Autologous Apheresis RBC Allogeneic Apheresis RBC Total Supply on TestingRejected OtherRejected for Reasons 38 10 Available Supply 7 - 48 3 7 0.4 0 1,835 21 0.1 1,856 WB for distribution as Whole Blood 49 13.7 100 units (all groups)RBC 4 Group O-positive 85 1967 5 –52.0Group O-negative 5 1 613 13,063 6 90 527 1 13,590 WB/RBC Distributed 4 <0.0001 75.0 0.7 207 –5.6 12,365 506 100.0 <0.0001 16 12,871 629 15,468 197 Allogeneic WB/RBC ( including directed) <1.0 48 0.001 94.7 11,850 98 32 Autologous 526 14,352 12,376 4.6 9 1,018 –12.1Total Transfusions199 – 5,084 72 99.9 6,109 <0.0001 5,156 1,245 14,982 –8.2 –10.318 –38.21,263 83 11,856 <0.0001 23 527 0.008 41.7 12,383 <0.0001 10.2 –17.4 1,358 5,648 200 <0.0001 100.0 –7.0 –8.7 15,030 <0.0001 <0.0001 9 - 13 13 –17.6 6,122 3 0.2 <0.0001 31 6,131 –58.1312 100.0 <0.0001 6,615 –7.3 0.036 WB Allogenic (includingdirected) 11,183 496 11,679 187 85.9 13,397 –12.8 Table 4-1. Estimated 2013 Colle data. 2011 from *Significantly different (WB) and Red Blood Cells (RBCs) (expressed in thousands of units) Distribution Transfusion Collection

Blood Collected and Produced in AABB Member Facilities: Trends in US Collections 13 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT in the number of blood cen- Non-RBC Components The split rate is the average ters can be attributed to a number of units produced large multiregional blood Non-RBC components col- from an apheresis proce- center that reported in lected or produced include dure; each unit must con- aggregate in 2013 but by  11 apheresis and WBD plate- tain a minimum of 3 10 region in 2011, as well as to lets, apheresis and WBD platelets. In 2013, the over- blood center mergers and plasma, cryoprecipitated all split rate was 1.8 apher- closures. Among the institu- AHF, and granulocytes. The esis platelet units per tions that reported RBC total number of non-RBC procedure. There was a sta- apheresis collections components distributed for tistically significant 4.2% (unweighted data), the transfusion in 2013 was decrease in the number of mean number of units col- 7,646,000 (WBD platelets apheresis platelet compo- lected by blood centers was counted as individual con- nents distributed by AABB 25,000 (vs. 16,000 in  centrates, not as apheresis- blood collectors in 2013 2011), and by hospitals it equivalent units), a decline (2 million units ±50,000). was 410 (vs. 745 in 2011). of 5.6% from 2011 distribu- tions. In 2013, US blood centers distributed 819,000 WBD RBC Distributions platelet concentrate units Platelets (164,000 apheresis equiva- There were 12.9 million lent units) individually and (±197,000) WB/RBC units in pools. Compared with An estimated 1,241,000 available for distribution in 643,000 WBD platelets dis- platelet apheresis proce- 2013; 12.4 million RBCs tributed in 2011, there was dures were completed by (±199,000) were distributed a significant 27.4% US-based AABB member by AABB member blood increase in WBD platelet blood collectors in 2013. collectors in 2013, com- distributions. This large This was 4.3% fewer proce- pared with 15.0 million dis- increase represents a real dures than reported by tributed in 2011. Of these, change in the pattern seen AABB members in 2011 5.2 million (±83,000; 41.7%) in previous surveys of (Figure 4-2). These proce- were Group O-positive decreasing WBD platelets dures yielded 2.2 million units, compared to 37.6% being produced and distrib- apheresis units (±55,000); (5.6 million) Group O-posi- uted nationally. 98.3% of which were col- tive units distributed in lected in plasma. Another 2011. There were 1.3 mil- 38,000 were suspended in In 2013, the average num- lion (±23,000) Group O- platelet additive solution ber of WBD platelet con- negative units distributed, (PAS). Blood centers col- centrates reported to be representing 10.2% of dis- lected 94.8% of apheresis pooled together for transfu- tributions in 2013. In 2011, platelets. This 2.5% sion was 5 units. This is Group O-negative units decrease in apheresis plate- consistent with the pattern accounted for 9.0% of total lets was not statistically sig- seen in previous surveys. distributions. nificant. For comparison with apher- esis platelets, it is assumed that 5 WBD platelet con-

14 Blood Collected and Produced in AABB Member Facilities: Trends in US Collections THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

1,400,000

1,200,000

1,000,000

800,000 2011

2013 600,000

400,000

200,000

0 RBC Apheresis Platelet Apheresis Plasmapheresis

Figure 4-2. Apheresis procedures by primary component.

centrates are equivalent to or produced in 2011 tals produced another 1 unit of apheresis platelets. (p<0.0001). This included 283,000 units. Compared several types of WBD to 2011, there was a 29.9% There were a total of plasma: fresh frozen plasma decline in the number of 2,166,000 platelet aphere- (FFP), plasma frozen within plasma units produced from sis equivalent units distrib- 24 hours after phlebotomy WB. The largest reported uted in 2013, 2.4% fewer (PF24), cryoprecipitate change was in the produc- than were distributed in reduced plasma (plasma tion of liquid plasma, where 2011—of this total 92.4% cryoprecipitate reduced), only 79,000 units were pro- were apheresis platelets liquid plasma, and other duced in 2013, compared units (Table 4-2, Figure 4-3). plasma for transfusion, as to 550,000 in 2011 (Figure well as plasma from apher- 4-4). esis collections. The total Plasma amount of plasma collected A total of 81,000 plasma- by apheresis or produced pheresis procedures were from WB represents a reported, generating an esti- A total of 4.3 million units decrease of 26.0% from mated 266,000 units of (±118,000) of plasma were 2011. Blood centers pro- apheresis plasma for trans- collected or produced for duced 93.4% of the plasma fusion and 3500 units of transfusion; this is 26.0% (4 million units) and hospi- source plasma. Multicom- fewer units than collected

Blood Collected and Produced in AABB Member Facilities: Trends in US Collections 15 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT % Change % Change 2011-2013 p-value  ns. 2011 Total onding facilities. 51 2,219 –2.4 0.249 104 1,019 12.2 0.112 55 2,283 –2.5 0.078 * * * as reported by the resp by the reported as 2013 Total CI ±95% Combined produced from whole blood donatio from whole blood produced BB US Member Blood Centers and Hospitals for Non-RBC ighted average units per pool 0 167 0 167(835) 1,054 53 1,054*(581) 116 30.7 132 0.142 634 66.2 <0.0001 154 9 164(819)*1,218 9 129(643)117 27.1 1,335* <0.0001 70867 54.0 <0.0001 Blood Blood ndividual platelet concentrates platelet ndividual Centers Hospitals † as individual units using we using units as individual ‡ usion 1,908 94 2,002* 50 2,090 –4.2 0.015  † thousands of units) rs in parenthesis represent i parenthesis represent rs in and Produced 2,112 114 2,226 r Transfusion 2,062 103 2,166 ncentrates Distributed ets Concentrates ‡ Apheresis Apheresis equivalent units; numbe pools expressed and units individual Includes Components (expressed in Table 4-2. Estimated 2013 Collection and Transfusion by AA different from 2011 data. *Significantly † ‡ Plasma Collected or ProducedPlasma Collected or Plasma Distributed for TransfusionCryoprecipitate Distributed for Transfusion 3,995 3,286 283 201 4,278* 3,488* 118 76 5,784 4,495 –26.0 –22.4 <0.0001 <0.0001 Total Platelets Distributed fo Total Platelets Transfusions Apheresis Platelets 0 1,143 1,143 WB-Derived Platel Total Platelets TransfusedPlasmaCryoprecipitate 0 1,310 1,310* 1 121 1,796 1,135 1,797* 15.4 129 0.0423 1,995 –9.9 0.036 Apheresis Platelets Distributed for Transf WB-Derived Platelets Co Collection/Production Apheresis Platelets Collected

16 Blood Collected and Produced in AABB Member Facilities: Trends in US Collections THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

100%

90%

80%

94.2% 92.4% 70% 83.7% 82.0%

60% 68.6% 67.7%

50%

40%

30%

20%

10%

0% 2001 2004 2006 2008 2011 2013

Apheresis Platelets WBD Platelets

Figure 4-3. Apheresis platelets as percent of total platelets produced.

ponent apheresis proce- duced increased over that fraction (0.1%) of this total. dures resulted in the in 2011, when 77.7% of Overall, this was a 14.6% collection of another 4,495,000 units were dis- decrease from 2011 levels. 266,000 units of apheresis tributed, suggesting plasma (Figure 4-4). Over- improved efficiency of all, plasma for transfusion blood collectors. In all, Cryoprecipitate collected by apheresis 340,000 units of group AB increased by 21.2% over plasma were distributed. In 2013, the average cryo- the amount collected in This was an increase of precipitated AHF pool size 2011. However, apheresis 16.5% from 2011 and rep- reported by blood collec- plasma collections com- resents 9.7% of the plasma tors that distributed pools prised only 12.4% of distributed for transfusion. was 5.5 individual units. An plasma for transfusion. estimate based on a There were 7 million units weighted average reported Of the plasma produced, of plasma distributed for pool size of the individual 81.5%, or 3.5 million units further manufacture, with cryoprecipitate units distrib- (±76,000), were distributed 98.1% from blood centers. uted by each collector for transfusion. While the Source plasma, (ie, plasma totaled 1,335,000 individ- number of units distributed collected by apheresis and ual units distributed singly decreased by 22.4%, the intended for further manu- or in pools by AABB mem- proportion of plasma dis- facture into derivatives), ber facilities. This was a sig- tributed to plasma pro- comprised only a small nificant increase of 54.0%

Blood Collected and Produced in AABB Member Facilities: Trends in US Collections 17 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

3000000

2500000

2000000

1500000

2011 AABB 2013 AABB

1000000

500000

0 FFP from WB PF24 from WB Plasmapheresis Other apheresis Cryoreduced Liquid plasma Other plasma plasma from WB for transfusion

Figure 4-4. Sources of plasma for transfusion.

over that distributed in Blood Center Inventory for task force and HHS use. 2011 (867,000 units; The organizations that sub- p<0.0001). In the course of delibera- mit data to AABB are Amer- tions on how to communi- ica’s Blood Centers, the cate the status of the blood American Red Cross, and Granulocytes supply to US Health and Blood Centers of America. Human Services (HHS) dur- The data are aggregated and disseminated through There were 1800 granulo- ing disasters, the AABB the AABB Research Depart- cyte units distributed; these Inter-organizational Task ment. The first complete were prepared from both Force on Domestic Disas- year for these supply esti- apheresis and WB buffy ters and Acts of Terrorism mates was 2008. Figure 4-5 coat units. This is a 19.7% settled on a simple quanti- indicates the overall decrease from the amount tative approach that reports national days of Type O red distributed in 2011. Blood the US blood center on- cell supply throughout the centers reported 80.9% of shelf blood supply in days 2013 calendar year com- this total. of available supply. AABB has been collecting and pared with the 2011 calen- reporting on supply daily dar year. These availability

18 Blood Collected and Produced in AABB Member Facilities: Trends in US Collections THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

Figure 4-5. Days of Group O supply 2011 and 2013.

data do not represent possi- fairly constant; the supply and 4 days of supply. How- ble geographic differences of O-positive units was ever, the O-positive supply in availability; however, comparatively high for the followed a cyclical pattern blood can be moved first part of the year, then with peaks in January and quickly from one location dropped mid-year and only May and with more robust to another through various recovered partially by year days of available supply supply networks. In 2011, end. In 2013, available O- throughout the year than the available supply of O- negative units remained reported in 2011. negative units remained fairly constant between 2

Blood Collected and Produced in AABB Member Facilities: Trends in US Collections 19

THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

5. Blood Transfused by US-Based AABB Member Hospitals

WB and RBCs Transfused transfusion (p=0.036). The units collected that were number of allogeneic units transfused, as in previous transfused, including units The 2013 AABB Blood Sur- surveys. Comparison of the directed to a specific vey collected data on trans- raw (unweighted) data for patient, was 7.1% less than fusions performed by AABB 181 AABB hospitals report- reported in 2011. Of the member institutions. In this ing to both 2011 and 2013 red cell units transfused report, comparisons have surveys indicates a “same among hospitals reporting been made between these store” average reduction in transfusion totals for Group 2013 data and the AABB WB/RBCs transfusion of O components, 40.5% institutional member sub- 6.1%, a difference consis- (1,125,000 units) were set of the 2011 NBCUS and tent with the overall Group O-positive and 9.6% as such, may not be gener- decrease in transfusion of (266,000 units) were Group alizable to all US institu- 7.3% described above. O-negative. tions involved in blood transfusion. In certain cir- cumstances, extrapolations The number of autologous Pediatric Transfusions to the entire US hospital transfusions continued to community are made, but decline significantly In 2013, transfusions to these should be treated with (p<0.0001) in 2013. There pediatric patients were caution as they are not were 58.1% fewer autolo- assessed by asking for the made with weighted, statis- gous units transfused than total number of transfu- tically valid sampling tech- in 2011 (13,000 units sions, including units, ali- niques. ±3000 in 2013 vs. 31,000 quots, and/or syringes, to in 2011). Transfusion of the pediatric or neonatal WB, always a small contri- AABB member facility population.* When quanti- bution to the total transfu- transfusions of WB and ties reported by AABB sion picture, decreased by RBCs, including autolo- member hospitals in 2013 15.4% in 2013 to approxi- gous transfusions, totaled were compared to those mately 13,000 units. 6,131,000 units (±312,000 reported by AABB member units) in 2013 (Table 4-1). hospitals in 2011 for WB/ As compared to the 2011 The use of the AABB-only RBCs and platelets, there AABB hospital cohort, this hospital cohort restricts the ability to draw conclusions was a statistically signifi- *Pediatric and neonatal popula- cant 7.3% drop in WB/RBC about the proportion of tions, as defined by the institu- tion.

Blood Transfused by US-Based AABB Member Hospitals 21 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT were decreases of 4.1% and received 1,638,514 units, An estimated total of 8.7%, respectively (Table 5- or 2.72 units per recipient, 1,310,000 platelet units 1). However, there was a very similar to the rate of (±121,000) were transfused 55.4% increase in the 2.75 units per recipient by US AABB member hos- amount of plasma reported reported by all hospitals in pitals in 2013, a significant transfused to this popula- 2011. Autologous recipi- increase of 15.4% in com- tion (87,000 transfusions in ents received an average of parison with transfusions by 2013) when compared to 1.4 units per transfusion (vs. US member hospitals in that transfused in 2011 1.3 units in 2011). For pedi- 2011 [(p=0.04), Figure 5-1]. (56,000 transfusions). Pedi- atric transfusion recipients, The transfusion of apheresis atric transfusions comprised the ratio was 3.0 units/ platelets increased by 5.0% of all WB/RBC trans- aliquots/syringes transfused 12.2%, from 1,019,000 to fusions, 9.4% of all platelet per recipient in 2013, an 1,143,000 units. transfusions, and 4.9% of increase from the reported all plasma transfusions in rate of 1.9 units per recipi- In this report, as described 2013. ent reported by hospitals in in Chapter 4, platelets are 2011. reported using apheresis equivalents. For compari- Transfusion Recipients son with the transfusion of Non-RBC Components apheresis platelets, it is The 2013 AABB Blood Sur- Transfused assumed that 5 WBD plate- vey captured the number of let concentrates are equiva- recipients of transfused WB/ AABB member hospital lent to 1 unit of apheresis RBCs. Based on estimates for non-red cell platelets. There were unweighted data from 316 components transfused in 167,000 apheresis-equiva- facilities reporting numbers 2013 and comparisons with lent units of WBD platelets of transfusion recipients, AABB member transfusions (835,000 individual WBD there were 602,955 recipi- in 2011 are presented in platelets units) transfused in ents of allogeneic WB/RBC Table 4-2. 2013, an increase of units; these patients 30.7%. Of these WBD

Table 5-1. Pediatric Transfusions by US Blood Centers and Hospitals in 2013 (expressed in thousands of units)

2013 AABB Hospitals 2011 AABB Hospitals

Total Number of Total Number of Transfusions Transfusions (units, (units, aliquots, and/or aliquots, and/or Pediatric Transfusions syringes transfused) syringes transfused) % Change WB/RBCs 306 319 –4.1 Platelets 105 115 –8.7 Plasma 87 56 55.4

22 Blood Transfused by US-Based AABB Member Hospitals THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

Figure 5-1. Trends in platelet transfusion, 1997-2013.

units, 74,000 platelet con- minor discrepancy is within representation of those that centrates were transfused as the range of reporting error transfused WBD platelets individual units and for a survey this size, a fur- among the reporting AABB 761,000 platelet concen- ther concern relates to the hospitals; or there were trates were transfused in fact that responding AABB inaccuracies in tabulating pools. There was a change hospitals represent approxi- and/or reporting WBD in the ratio of apheresis mately 48% of all transfus- platelet transfusions due to concentrates to WBD plate- ing US hospitals; thus the confusion between the use let concentrates from 8.8 number of transfused WBD of individual and pooled apheresis units transfused platelets substantially concentrates. Of interest, a for every pool of WBD con- exceeds that to be expected similar proportional dis- centrates in 2011 to 6.8 from AABB member hospi- crepancy was reported for apheresis units for every tals. In contrast, the propor- cryoprecipitated AHF pool of WBD platelets in tion of distributed apheresis (which also may be trans- 2013. platelets transfused by fused as individual units or AABB member hospitals is in pools) in that AABB There were approximately much lower (57.1%) and member hospitals trans- 819,000 WBD platelet con- more in keeping with fused a disproportionally centrates (164,000 aphere- expectations. These WBD high percentage (80%) of sis equivalents) distributed platelet findings were con- distributed units. and 835,000 reported firmed during the data (167,000 apheresis equiva- cleaning process and may The combined total of lents) transfused by AABB have one of two explana- WBD and apheresis plasma hospitals. While this overall tions: there was an unequal transfused in AABB member

Blood Transfused by US-Based AABB Member Hospitals 23 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

hospitals was 1,797,000 Hospitals were asked a sep- Of all plasma transfused, units (±129,000), signifi- arate question about the 71,000 units of Group AB cantly less (9.9%, p=0.04) percentage of plasma plasma were reported trans- than the amount transfused issued as Thawed Plasma. fused in 2013. This is a in 2011 (1,995,000 units). Thawed Plasma is derived 9.2% increase over that Reporting institutions indi- from FFP or PF24 and transfused by AABB hospi- cated the amounts of the thawed at 30o C to 37o C, tals in 2011 (65,000 units) various types of plasma then maintained until trans- and represents approxi- transfused as shown in Fig- fusion at 1o C to 6o C for up mately 10.5% of plasma ure 5-2. There was a to 5 days. The average per- transfused for those facili- marked increase in the centage of plasma issued as ties that reported both transfusion of liquid plasma thawed among member amounts of AB plasma and among AABB member hos- hospitals was 53.0% com- all plasma transfused. pitals. There were 24,000 pared with 24.5% in AABB units of liquid plasma trans- member hospitals in 2011. When asked how institutions fused, compared with only The average percentage of routinely order plasma 2000 units of liquid plasma plasma issues as Thawed transfusions to non- reported transfused in Plasma was positively asso- pediatric patients, most 2011. ciated with the annual transfusing facilities number of surgeries. (Figure (56.3%) reported routinely 5-3).

2,000

1,800

1,600

1,400

1,200

1,000

2011 AABB hospitals 2013 AABB hospitals 800

600 Units transfused in thousands in transfused Units 400

200

0 Plasma (including Plasma Plasma-jumbo Liquid Plasma Pediatric Plasma Group AB Plasma FFP, PF24, and Cryoprecipitate Thawed Plasma) Reduced

Figure 5-2. Types of plasma transfused in 2011-2013.

24 Blood Transfused by US-Based AABB Member Hospitals THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

70

2011 AABB 2013 65

60

54 53

50 49 47

41

40 37

34

30

20 20

Percent of Plasma Transfusions Transfusions of Plasma Percent 14 11 9 10

0 100-999 surgeries/year 1000-1399 surgeries/year 1400-2399 surgeries/year 2400-4999 surgeries/year 5000-7999 surgeries/year шϴϬϬϬƐƵƌŐĞƌŝĞƐͬLJĞĂƌ

Figure 5-3. Percent of plasma transfused as thawed plasma (2013 average 53%).

transfusing plasma based tated AHF transfused in of massive transfusion pro- on perceived level of coag- 2013. This is a large and tocols. ulation factor deficiency or significant increase (66.2%) degree of bleeding (com- over the amount transfused Transfusion of granulocytes, pared to 62.0% in 2011). In in 2011 (634,000 units). Of prepared from both aphere- 5.7% of the transfusing the total, 164,000 units of sis and WB buffy coat units, facilities, the dosage was cryoprecipitate was trans- decreased by 3.7%, from based on patient weight fused as individual concen- 2062 units in 2011 to 1985 (6.6% reported used this trates and 890,000 units in 2013. However, methodology in 2011). cryoprecipitate concen- because these numbers are Other facilities reported trates were made into pools so small, estimates should transfusing a standard num- and transfused as such. not be regarded as com- ber of units regardless of the pletely reliable, as illus- patient’s weight (11.3% vs. Among the reasons trated by the data indicating 12.4% in 2011) and 26.7% reported for the large that more units were trans- reported that the number of increases in WBD platelet fused (1985 units) than units transfused was not and cryoprecipitate transfu- were distributed (1800 consistent with any of the sion are the addition of crit- units). above (10.0% in 2011). ical care beds, acquisition of other facilities, increases A total of approximately There were approximately in liver and solid organ 20,000 specialized prod- 1,054,000 units or unit transplants, and the impact ucts (HLA matched or equivalents of cryoprecipi-

Blood Transfused by US-Based AABB Member Hospitals 25 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT crossmatched) were trans- There were 18,000 autolo- positive units distributed, fused to alloimmunized gous units outdated in 2013 only 1.1% were outdated patients in 2013. (32.7% of AABB member and of the Group O-nega- autologous collections), tive units distributed, only The total number of units of compared to 20.5% in 0.7% were outdated. all components transfused 2011. In addition, AABB by AABB member institu- hospitals reported 2000 There were 220,000 apher- tions in the United States in autologous units as wasted. esis platelets units outdated 2013, both RBC and non- in 2013. Of the apheresis RBC components, was A much smaller amount of platelets distributed, 11.0% 10,962,000, a 1.1% WB for transfusion as WB were outdated compared to increase over the AABB was distributed in 2013 the 11.5% outdated in member transfusions compared with 2011. 2011. There were 16,000 (10,846,000 units or unit Approximately 12.0% of units reported wasted or equivalents) from 2011.* 6400 WB units distributed discarded for non-test This increase was predomi- were outdated (11.8% in result-related reasons. nantly due to the changes 2011). in transfusion of WBD WBD platelet concentrates platelets and cryoprecipi- Approximately 3.1% of accounted for 20.8% of tate. allogeneic red cell dona- total outdated components, tions were reported as out- or 194,000 units. This rep- dated in 2013, compared to resents 23.7% of the WBD Outdated Units 1.7% in 2011. As shown in platelets distributed in Table 5-2, outdated WB/ 2013, whereas in 2011, The AABB member esti- RBCs accounted for 3.4% 34.7% of WBD platelets mate for the total number of of all WB/RBC units distrib- that were distributed were units of WB and all compo- uted in 2013. An additional subsequently outdated. In nents outdated by blood 695,000 units were addition to outdates, 2.8% centers and hospitals in reported wasted or dis- (23,000 units) of distrib- 2013 was 932,000 units. carded for non-test result- uted WBD platelets were Outdated WB and RBCs related reasons. reported as wasted, with the totaled 421,000 units reasons given as breakage, (±14,000 units). This As in 2011, the current sur- out of temperature, and included autologous and vey inquired about outdates non-outdated units that allogeneic RBCs as well  of Group O-positive and O- could not be transfused. as WB distributed for negative RBC units (Figure transfusion. 5-4). In 2013, they Due to their limited shelf- accounted for 16.5% of the life, apheresis platelets and total outdated allogeneic WBD platelets accounted RBCs (14.2% O-positive, for nearly 44.4% of out- *The total transfusions are from 2011 AABB facilities. The total 2.3% O-negative), higher dated units in 2013. units include WB/RBCs, WBD than 11.3 % in 2011 (AABB platelets, apheresis platelets, facilities). While 3.1% of all Outdated plasma of all plasma, cryoprecipitated AHF, allogeneic red cells were types totaled 62,000 units, and granulocytes. outdated, of the Group O- 1.8% of the plasma units

26 Blood Transfused by US-Based AABB Member Hospitals THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT and pools)and Granulocytes concentrates Cryoprecipitate Cryoprecipitate (both individual individual (both s of Each Component Distributed for pools) Plasma 819,000 3,488,000 1,335,000 2,000 Concentrates (both individual individual (both concentrates and WB-Derived Platelets Platelets WB-Derived Platelets Apheresis rcentage of the Total Number Unit  WB/RBCs 6,131,000 1,143,000 835,000 1,797,000 1,054,000 2,000 er hospitals included. er † Transfusion in 2013 Table 5-2. Outdated Components as a Pe AABB member hospitals only. Outdated TotalOutdated Distributed421,000 *Blood centers and AABB memb † 12,383,000220,000 2,002,000 194,00062,000 35,000 74 Overall Percentage OutdatedPercentage Overall Wasted*Reported 3.4 Transfused 11.0 695,000 23.7 16,000 1.8 23,0002.6 3.7 47,000 67,000 39

Blood Transfused by US-Based AABB Member Hospitals 27 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

O-negative 2.3%

O-positive 14.2%

Other groups and types 83.5%

Figure 5-4. Percentage of Group O allogeneic RBC outdates.

distributed for transfusion. were outdated, 2.6% of the dated. Some components This was 6.7% of total out- cryoprecipitated AHF dis- were included in both the dates. There were 47,000 tributed. An additional outdated and wasted totals; units reported wasted or 5.0% (67,000 units) were therefore, it may appear discarded. reported wasted. that more products were available than were There were 35,000 units of Lastly, 3.7% of granulo- reported distributed. cryoprecipitated AHF that cytes distributed were out-

28 Blood Transfused by US-Based AABB Member Hospitals THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

6. Patient Blood Management (PBM) in AABB Member Facilities

This is the second survey of blood transfusions, poten- the PBM section. Their PBM activities by hospitals tially avoiding transfusion- detailed results are not and blood centers. A PBM associated complications included here. section was included as and reducing health care Chapter 5 of the 2011 costs, while ensuring that In 2011, 591 AABB member NBCUS report. Data were blood components are facilities, including 529 not weighted or imputed available for the patients (89.5%) hospitals, responded and are summarized here as who need them. The ques- to the initial survey ques- reported. Denominators tions in the PBM section of tion that asks whether the reflect the number of facili- the AABB Blood Survey facility had a PBM program ties that answered each of were designed to build on in 2011. In both surveys, the questions. the 2011 NBCUS questions many facilities that did not and address current topics report having a PBM pro- This report reflects changes of interest. gram provided responses to in the field of transfusion some of the questions in the since 2011 among AABB In 2013, 607 US AABB PBM section. member hospitals. PBM is member institutions an evidence-based, multi- responded to the initial sur- Many combinations of disciplinary approach to vey question that asks medical professionals were optimizing the care of whether the facility had a reported to coordinate PBM patients who might need  PBM program in 2013, 547 programs. In 2013, the a blood transfusion. It (90.1%) were hospitals. Of majority of PBM programs encompasses all aspects of the hospitals that responded (90.8%) were coordinated patient evaluation and clin- to the question of whether by a medical director alone ical management surround- their institution had a PBM or as part of a team.* The ing the transfusion program in 2013, 37.8% medical specialty of these decision-making process, responded affirmatively. directors was most often including the application of reported to be pathology appropriate transfusion In addition, six non-US (33.5%), transfusion medi- indications, as well as mini- AABB member hospitals cine (7.4%), and anesthesi- mization of blood loss and reported having PBM pro- ology (13.8%). Other optimization of patient red grams. Eight other non-US cell mass. As a consequence facilities provided responses *Of hospitals responding to this of better management, to some of the questions in question. patients may require fewer

Patient Blood Management (PBM) in AABB Member Facilities 29 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT specialties coordinating 30.9% reported a single with hospitals reporting the PBM programs included PBMC on staff, another 52.7% use of guidelines from surgery, hematology, and in reported at least one part- AABB (76.4%), College of a few facilities, the hospi- time PBMC. All hospitals American Pathologists tal’s chief medical officer. A with PBMCs reported them (29.2%), American Society total of 27.1% of hospitals to be hospital employees. of Anesthesiologists (2.8%), with PBM programs American Red Cross reported nursing staff In 2013, 41.9% of respond- (8.6%), and others (17.7%). involvement in the coordi- ing hospitals reported par- Other guidelines used nation of the program, and ticipating in performance included recommendations 42.0% reported involve- benchmarking programs from the US Food and Drug ment of other non-nursing relating to transfusion medi- Administration (FDA), the staff including blood bank cine (43.7% in 2011).* National Institutes of Health staff, risk management staff, (NIH), The Joint Commis- health care improvement sion, the New York State staff, transfusion commit- Sixty-six percent of hospi- Department of Health, tees, blood utilization com- tals reported providing for- internal transfusion com- mittees, quality assurance, mal transfusion training mittees, and/or multiple and patient safety officers. and/or education to their sources of evidence-based staff, compared to the practice literature. 70.1% reported in 2011. There were 74 AABB mem- This training was required ber hospitals (13.5% of all for anesthesiologists and Of the institutions with reporting hospitals) that other physicians who trans- transfusion guidelines, reported having transfusion fuse blood at 37.5% of 77.5% reported that their safety officers (TSOs); of these facilities. Of hospitals hospital guidelines had these, 51.4% reported a reporting PBM programs in been incorporated into PBM program in 2013. 2013, 42.0% provided for- paper or electronic order- Most hospitals (35.1%) with mal PBM training to physi- sets. Sixty-five percent TSOs reported having one cians and 39.6% provided reported that their comput- full-time TSO on staff and training to nurses. erized physician order entry another 55.4% reported (CPOE) included transfu- having at least one part- sion guidelines. Of these, time TSO. More hospitals The use of transfusion 53.8% have guidelines that reported that TSOs were guidelines was reported by are a “hard stop,” where the hospital employees (57.4%) 78.2% of reporting hospi- physician/provider must than blood center tals. This was a significant select a choice before the employees. decline from the number order is accepted. Only reported in 2011 (92.8%; 45.9% of these hospital p<0.001). Of the hospitals There were 55 hospitals CPOE systems included  using transfusion guide- an algorithm or clinical (10.1% of all reporting hos- lines, 89.5% reported using pitals) that reported having decision support that warns one or more of the national or alerts the physician/pro- a PBM coordinator (PBMC); guidelines (Figure 6-1), of these, 86.6% reported vider if they are ordering a having a PBM program. Of blood product outside of hospitals with PBMC staff, *Of hospitals responding to this guidelines. question.

30 Patient Blood Management (PBM) in AABB Member Facilities THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

Figure 6-1. Use of national transfusion guidelines, 2011-2013.

Most reporting hospitals In 2013, only 28.4% of hos- manage a patient’s anemia (71.9%) required that a pitals reported the evalua- before surgery. physician document the tion of factors predictive of reason or clinical justifica- pre- and postoperative ane- tion for transfusion in the mia for patients facing elec- PBM Interventions medical record based on tive surgical procedures guidelines developed by associated with a high like- Many hospitals, both those the hospital transfusion or lihood of blood loss, a drop with and without PBM pro- quality committee. As com- of 35.2% from that reported grams, have formally imple- pared to 2011, in 2013 in 2011 (63.6%). Twelve mented interventions to fewer hospitals (52.8% vs. percent of hospitals reduce the need for transfu- 68.5%) required that rele- reported evaluating all of sion. In 2013, 199 AABB vant pre-transfusion labora- these patients and 16.9% member hospitals reported tory results be documented reported such evaluations implementation of at least for non-emergent transfu- for only some types of sur- one preoperative interven- sions. Fifty-two percent of gery (eg, cardiac). Of the tion. The specific interven- hospitals reported promot- hospitals conducting these tions are displayed in ing the use of single RBC evaluations, 34.3% Figure 6-2. Laboratory transfusion orders in 2013. reported having a program assessment for anemia was or formal service in place to

Patient Blood Management (PBM) in AABB Member Facilities 31 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT  terventions implemented, 2013 (n=199). Figure 6-2. Preoperative PBM in

32 Patient Blood Management (PBM) in AABB Member Facilities THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT the most common interven- program), parenteral (IV) ative blood recovery tion reported by 66.3% of iron supplementation (n=206 or 91.2%), 55.3% hospitals, and of these 54.5 (27.6% of hospitals report- of which did not report hav- % were hospitals that had a ing preoperative interven- ing a PBM program. Among PBM program. Other pre- tions), erythropoietin the 67 hospitals (29.6%) operative interventions administration (25.1% of reporting implementation of implemented included the those reporting preopera- acute normovolemic hemo- use of the clinical service tive interventions) and pre- dilution, 52.2% did not for assessment for anemia operative autologous blood have a PBM program in (37.2% of those reporting donation (40.7% of those place. Use of topical/sys- preoperative interventions), reporting preoperative temic hemostatic agents the use of the clinical ser- interventions, 54.3% of was reported by 108 hospi- vice for assessment for these also reported having a tals (47.8%) and 50.9% of bleeding risk (27.6% of PBM program). them reported having a those reporting preopera- PBM program. tive interventions, 58.2% of As shown in Figure 6-3, which also reported having 226 hospitals reported In 2013, 203 hospitals a PBM program), oral iron implementation of intraop- reported the use of postop- supplementation (39.7% of erative interventions in erative interventions. Figure those reporting preopera- 2013. Of these, the majority 6-4 shows that the most tive interventions, 70.9% of had implemented intraoper- common postoperative which also reported a PBM

Figure 6-3. Intra-operative PBM interventions implemented, 2013 (n=226).

Patient Blood Management (PBM) in AABB Member Facilities 33 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

Figure 6-4. Postoperative PBM interventions implemented, 2013 (n=203). intervention reported was tion (eg, from chest tubes) metrics used by their insti- the restrictive use of postop- and re-administration tution to measure the suc- erative transfusion (72.9% (29.6% of those reporting cess of interventions or 148 of 203 respondents); postoperative PBM activi- intended to improve PBM 68.2% of these hospitals ties). (Figure 6-5). The two most reported having a PBM pro- common measures (which gram. Other postoperative are similar) reported by hos- interventions frequently Measure of Intervention pitals in 2013 were total reported by hospitals Success RBCs, platelets, and plasma included the reduction in transfused (71.5%) and the the frequency of blood total number of compo- draws for lab testing (39.9% Fifty-six percent of hospitals nents transfused (57.3%); of hospitals reporting post- that completed some or all the latter was the most operative interventions) and of the PBM questions pro- common measure reported postoperative blood collec- vided information on the by hospitals in 2011. Other

34 Patient Blood Management (PBM) in AABB Member Facilities THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

Figure 6-5. Methods hospitals use to measure the success of implementations to improve PBM. measures were the number RBC units per 1000 patient- Informed Consent of RBC units per patient days (5.5%), adjusted transfused (41.7%), and the patient-days (4.9%), and Most AABB member hospi- percentage of patients patient encounters (1.6%). tals required the ordering transfused per hospital Measures reported in the provider to obtain and doc- admission (19.4%). “Other” category (17.2%), ument informed consent for  included the crossmatch- transfusion in 2013 All other measures were to-transfusion ratio; reviews (91.4%). Among the hospi- reported to be used by of clinical waste, blood uti- tals requiring transfusion 17.8% or fewer hospitals. lization, and other audits; informed consent, 86.6% These included transfu- the percentage of patients have a separate consent sions per surgical case transfused per selected form for transfusion. Of (17.8%), transfused RBC ICD-9 code (ie, coronary these, 63.5% of hospitals units per 100 hospital artery bypass graft [CABG] reported that both the phy- admissions or discharges only or knee/hip replace- sician and patient sign the (13.9%), transfusions per ment); average blood com- informed consent, whereas medical/surgical admission ponents use per case 24.2% reported that the (9.1%, but previously (CABG only); and percent- informed consent form was reported by 23.3% of facili- age of transfusions that signed by the patient/repre- ties in 2011), transfused were deemed inappropriate. sentative only. A very small

Patient Blood Management (PBM) in AABB Member Facilities 35 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT number of hospitals (2.8%) determining when to trans- reported a policy to trans- reported that only the phy- fuse blood products. fuse between 8 and 9 g/dL sician signs the transfusion Although many hospitals and 41.6% reported poli- consent form. did not report having a cies to transfuse at more PBM program, 75.3% of than 9 g/dL. Of 272 hospi- hospitals responding to this tals reporting hemoglobin Type and Screen before part of the survey reported transfusion thresholds for Procedures having a policy for a hemo- the general outpatient pop- globin threshold for transfu- ulation/clinic, most poli- sion for hemodynamically cies were to transfuse at a In 2013, 82.9% of respond- stable, non-bleeding hemoglobin rate between 8 ing hospitals reported that patients. Figure 6-6 displays and 9 g/dL (46.3%) or they did not track the the AABB member survey between 7 and 8 g/dL percentage of “Type and responses when asked (44.1%). Screens” that were com- about specific aspects of pleted before the start of a their hemoglobin threshold surgical procedure. Of the policy. Of the 341 hospitals Platelets 86 hospitals that reported that reported thresholds for tracking that Type and the general inpatient popu- Of hospitals that responded Screens were completed lation, 55.1% reported a to questions about platelet prior to surgery, most policy to transfuse at a transfusion thresholds, (n=72) tracked all surgical threshold of between 7 and 73.1% reported having pol- cases and fewer tracked 8 g of hemoglobin per dL of icies for platelet transfusion selected high-blood loss blood. Among 282 hospi- thresholds. Table 6-1 dis- surgical cases (n=14). tals reporting thresholds for plays the AABB member Among those hospitals that surgical (postoperative) survey responses when did track this factor, two patients, 50.7% reported asked about platelet trans- thirds of the hospitals had that their policy was to fusion thresholds for spe- over 80 percent of patients transfuse at a threshold of cific patient populations. with a Type and Screen on between 8 and 9 g/dL. Of Among the 314 hospitals record before a surgical the 270 hospitals that that reported thresholds for procedure. reported thresholds for the the general, non-bleeding intensive care unit (ICU) inpatient population, patient populations (exclu- 62.4% of hospitals reported Transfusion Thresholds sive of acute coronary syn- a policy to transfuse plate- drome), 47.7% reported a lets at thresholds between Hemoglobin policy to transfuse between 5000 and 10,000 per 7 and 8 g/dL; another microliter (mcL) (n=196). For the 2013 survey year, 40.0% reported that their Of the 179 hospitals that questions were added to policy was to transfuse reported their thresholds for evaluate policies for the use between 8 and 9 g/dL. A inpatients with fever, of hemoglobin, platelet, total of 274 hospitals 61.5% reported a policy to International Normalized reported hemoglobin transfuse at a threshold Ratio (INR)/Partial thrombo- threshold policies for between 11,000 and plastin time (PTT), and patients with acute coro- 20,000 per mcL (n=110). fibrinogen thresholds for nary syndrome; 45.3% Of the 151 hospitals that

36 Patient Blood Management (PBM) in AABB Member Facilities THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

Figure 6-6. Reporting hospitals policy for a hemoglobin threshold for transfusion for hemody- namically stable non-bleeding patients. reported on inpatients fuse platelets at a threshold patients who had been receiving anticoagulants, between 21,000 and receiving antiplatelet medi- 51.7% reported transfusion 50,000 per mcL (n=215). cation. Of these, 62.2% thresholds between 11,000 Of 268 hospitals reporting reported thresholds of and 20,000 per mcL (n=78). thresholds for bleeding 21,000 to 50,000 per mcL Of hospitals that reported patients, 72.4% reported a (n=61). thresholds for patients with policy to transfuse platelets planned bedside invasive at a threshold of 21,000 to procedures (eg, line place- 50,000 per mcL (n=194). Plasma ment, [n=265]), 81.9% Among the 224 hospitals transfuse platelets at thresh- that reported thresholds for Of hospitals that responded olds between 21,000 and patients with central ner- to questions about plasma 50,000 per mcL (n=217). vous system (CNS) bleed- transfusion thresholds, Of the 309 hospitals that ing, 73.7% reported a 63.6% reported having an reported thresholds for policy to transfuse platelets INR/PTT threshold policy patients with planned oper- at thresholds of 80,000 per for the transfusion of ating room (OR) procedures mcL or greater (n=165). plasma. Figure 6-7 presents (eg, major surgery), 69.6% Only 98 hospitals reported the AABB member survey reported a policy to trans- thresholds for bleeding responses when asked

Patient Blood Management (PBM) in AABB Member Facilities 37 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT − − − −

− − (51,000 - 80,000/mcL) (>80,000/mcL) (21,000 - (21,000 50,000 /mcL) # Hospitals Responding (11,000 - (11,000 20,000/ mcL) 20,000/ 8 9 61 2 18 21 12 215 14 47 23 16 217 9 53 40 110 78 16 27 6 (5000 - (5000 10,000 /mcL) for Platelet Threshold Transfusion − − 2− 196 94 22 − − (<5000/mcL)  receiving antiplatelet medication receiving invasive procedure population Table 6-1. Reporting Hospitals Policy Bleeding patient patient Bleeding bleeding CNS with Patient has been Bleeding patient who 1 1 6 9 7 35 43 194 2 7 165 22 Patient with planned OR procedure Inpatient with fever Inpatient receiving anticoagulant bedside planned with Patient Platelet threshold for transfusionfor threshold Platelet General, non-bleeding inpatient N N N N N N

38 Patient Blood Management (PBM) in AABB Member Facilities THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

Figure 6-7. INR threshold for transfusion of plasma.

about specific aspects of fuse plasma at an INR of 1.0 thresholds, 27.6% reported their plasma transfusion to 1.5 and 44.3% reported that they transfuse plasma thresholds. Of the 237 hos- they transfuse plasma at an at a PPT of 30 seconds or pitals that reported thresh- INR of 1.6 to 2.0. Of the less and 25.3% reported olds for plasma transfusion 195 hospitals who reported transfusion of plasma when to the non-bleeding inpa- thresholds for plasma trans- the PTT is between 41 and tient population in prepara- fusion for bleeding patients, 50 seconds. PTT normal tion for emergent, imminent 56.4% reported a policy to values differ from lab to lab surgery or an invasive pro- transfuse plasma at an INR and often are considered cedure, 56.1% reported a of 1.6 to 2.0 and 40.0% not in seconds but as a ratio policy to transfuse plasma reported that they transfuse of patient to control value at an INR of 1.6 to 2.0 and plasma at an INR of 1.0  (with abnormal being 38.8% reported a policy to to 1.5. defined as >1.5 times the transfuse plasma at an INR control value). of 1.0 to 1.5. Among the PTT thresholds for bleeding 167 hospitals that reported patients are shown in Fig- When asked about policies thresholds for inpatients ure 6-8. Of the 174 hospi- for transfusion of cryopre- with CNS bleeding, 53.3% tals that reported PTT cipitate, 57.5% of hospitals reported a policy to trans-

Patient Blood Management (PBM) in AABB Member Facilities 39 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

Figure 6-8. Reported PTT threshold for bleeding patients (n=174).

reported having a policy on cipitate. Of these, 80.0% grams (mg) per dL and fibrinogen thresholds for reported a threshold of less 20.0% reported a threshold the transfusion of cryopre- than or equal to 100 milli- between 100 and 200 mg/dL.

Figure 6-9. Use of prothrombin complex concentrates (4-factor or 3-factor) by hospitals reporting a policy for their use.

40 Patient Blood Management (PBM) in AABB Member Facilities THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

Only 20.3% of hospitals rin in patients with intracra- hemorrhage, emergency reported having policies  nial bleeding, 79.3% surgery, and when the phy- for the use of 4-factor or  reported using PCCs for sician specifically ordered 3-factor prothrombin reversal of newer anticoag- its use. complex concentrates ulant drugs (eg, target- (PCCs). Figure 6-9 displays specific oral anticoagulants) Only 21.2% of responding the hospital use of PPC by in bleeding patients, and hospitals (n=116) reported patient population. Of hos- 28.9% reported use of PCCs having a policy on the use pitals with a policy for PCC for other indications, which of recombinant Factor VII. use, 98.9% reported using included liver failure, when PCCs for reversal of warfa- FFP is ineffective, massive

Patient Blood Management (PBM) in AABB Member Facilities 41

THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

7. Current Issues in Blood Collection and Donor Screening

In 2013, there were repeat donors.† These There were 2,035,000 13,975,000 individual vis- repeat allogeneic donors donors reported as deferred its, presentations, or regis- provided 7,415,000 dona- for various reasons (14.6% trations to donate blood at tions, the equivalent of 1.6 in 2013, compared to AABB member blood cen- donations per repeat alloge- 13.6% among AABB blood ters and hospital blood col- neic donor, compared with collectors in 2011). An lection sites; this was a 1.5 from AABB institutions additional 892,000 defer- 21.0% decrease from AABB in 2011. rals were reported for facilities reporting in 2011. which no specific deferral As reported in earlier sur- There were only 35,000 reason were identified. veys, most donors pre- autologous donors (ie, Donors were most com- sented at blood centers donors who successfully monly deferred for low (95.8%), and only 4.2% donated a unit intended for hemoglobin (51.8% of presented to donate at hos- themselves) reported by deferrals vs. 48.9% in pital donor centers. Of the AABB facilities in 2013. 2011), defined as not meet- registrations, there were This is 56.6% fewer autolo- ing FDA blood hemoglobin 6,847,000 allogeneic gous donors than reported level requirements for (directed and community in 2011. Facilities reported blood donation. As seen in combined) donors who suc- 55,000 units collected Figure 7-1, other categories cessfully gave blood at (manual and apheresis red for deferral included high- AABB facilities in 2013, cell collections combined), risk behavior associated compared to 9,127,000 a rate of 1.6 units per autol- with men who have sex allogeneic donors in 2011,* ogous donor. Autologous with other men (MSM; a 24.9% decrease. In 2013, donations are collected at a 0.4% of deferrals both there were 2,210,000 higher rate in hospital years), other high-risk (32.3%) first-time donors blood banks, with autolo- behavior (as identified on and 4,626,000 (67.6%) gous units representing the Donor History Ques- 1.9% of all units collected, tionnaire, or DHQ; 0.7% as compared to 0.3% in vs. 1.0% in 2011), prescrip- *In the past, many blood centers blood centers. tion drug use (2.8% vs. and hospitals reported that they 3.6% in 2011), tattoos/ were unable to specify which donors had directed their dona- piercings (3.4% vs. 3.9% in tions; therefore, in 2013, all alloge- 2011), specific foreign neic donations were grouped and travel (7.5% vs. 7.3% in specific questions about directed †Repeat donors as defined by the 2011), and other medical donation were eliminated. reporting facility.

Current Issues in Blood Collection and Donor Screening 43 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

Other 12.1% Tatoo/Piercing 3.4%

Travel 7.5%

High risk behavior (Other) 0.7%

High risk behavior (MSM) 0.4% Low hemoglobin 51.8%

Other medical reasons 20.0%

Prescription drug use 2.8% Low ferritin 1.2%

Figure 7-1. Donor deferrals 2013.

reasons (20.0% vs. 19.1% reported in 2011, were 8.8% of donations, were in 2011). Newly reported deferred for other reasons, collected from donors over on this survey were defer- which included low weight, the age of 65 years, an rals for low ferritin, which inadequate interdonation increase from 7.8% in amounted to 1.2% of defer- interval, being under the 2011. rals. High-risk behavior donation age, and lan- deferrals are intended to guage. AABB members collected reduce the risk of transmis- 1,533,000 units from sion of infectious diseases, In 2013, 2,758,000 units minority populations including HIV and hepati- were collected from donors (including African Ameri- tis viruses. Deferrals for aged 16 to 24. This number can, Asian, and/or His- other medical reasons may represents approximately panic). Although some include exposure to one-fifth of all allogeneic facilities were unable to human-derived growth hor- collections (20.4%), as it specifically categorize their mone, bovine insulin, hep- did in 2011. Of these dona- donors by race or ethnicity, atitis B immune globulin, tions, 1,333,000 were col- it is clear that donations unlicensed vaccines, or lected from high school- from minority populations those presenting with physi- aged donors (16 to 18 years continue to contribute sub- cal conditions or symptoms old), representing 9.9% of stantially to the nation’s that disqualify a person allogeneic donations. This blood supply (11.3%), the from donating blood. survey also determined that same percentage reported Another 11.9% of donors, a 1,194,000 donations, or for 2011. decrease from the 15.7%

44 Current Issues in Blood Collection and Donor Screening THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

Table 7-1. Donor Adverse Reaction Rate by AABB Facility and Procedure Type

Reaction Rate by Manual Reaction Rate for Automated Collection Procedure (%) Collection Procedures (%) Overall Rate (%)

2013 2011 2013 2011 2013 2011 Blood Center 0.13 0.14 0.13 0.10 0.13 0.13 Hospital 0.12 0.10 0.02 0.15 0.10 0.11 Total 0.13 0.13 0.12 0.11 0.13 0.13

Mobile blood drive sites for donor adverse reactions of severe AEs was 17,600/ were the source of have been established.* 14,005,000 collection pro- 9,634,000 units, or 70.9% Their widespread adoption cedures (0.13%) in 2013 of collected WB/RBC units. is anticipated and should and 21,000/15,721,000 Of these, approximately contribute to reporting con- collection procedures 12.2%, or 1,172,000 dona- sistency in the future. (0.13%) in 2011. These tions, were from automated rates were not statistically collections. As has been the This survey continues to different and indeed were historical pattern, in 2013, track donor hemovigilance almost identical. blood centers obtained the reports for the US donor greatest proportion of their population. For the pur- As in 2011, the rate of collections through mobile poses of this survey, severe severe AEs per unit col- blood drives (71.5%, an donor adverse events (AEs) lected in 2013 was the increase over the 67.6% in were defined as AEs occur- same statistically as that by 2011). Hospitals reported ring in donors that were procedure. Blood dona- higher use of mobile blood attributed to the donation tion, either through tradi- drives for collection in process, including major tional manual WB 2013, 56.3% of WB/RBC allergic reaction, loss of collection processes or collections compared to consciousness of a minute using automated proce- 50.9% in 2011. or more, loss of conscious- dures, is a very safe proce- ness with injury, and nerve dure, rarely resulting in irritation. Among AABB adverse consequences. Donor Hemovigilance members, 17,600 of these events were reported by In the past year, internation- collection organizations for ally harmonized definitions 2013 (Table 7-1). The rate

*http://www.aabb.org/research/ hemovigilance/Documents/ Donor-Standard-Definitions.pdf. Last accessed October 5, 2015.

Current Issues in Blood Collection and Donor Screening 45

THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

8. Current Issues in Blood Transfusion

US Collection Trends blood centers and hospitals the 12.1% decrease in the was 65.5 units per 1000 number of units collected. persons aged 16 to 64 in Furthermore, this collec- Figure 8-1 illustrates the 2013, compared with 76.2 tion rate is the lowest rate trends in the estimated rates units per 1000 persons of per unit of potential donor of WB/RBC collection and that age in 2011,† population reported since transfusion in the United a decline these large surveys have States from 1980 to 2013. of nearly 12.0%. Extrapo- been conducted in 1980. This year, the survey esti- lated to include non-AABB According to the number of mates AABB member rates hospital collections, the US donors reported in the 2013 of collection and transfu- collection rate is estimated survey year, only 3.3% of sion and compares these to be 67.2 units per 1000 the US population aged 16 with AABB member rates persons of donation age to 64 donated in 2013, a from 2011. (not shown in figure). This rate is slightly inflated since drop from the 4.5% of the we know that in 2013, 1.2 total age-eligible US popu- The rate of collection was million donations were col- lation reported to have calculated from the lected from donors over the donated in 2011. national estimate of total age of 65. Most of these allogeneic WB/RBC units donors are likely to be aged The rate of donation in the collected per 1000 donors 65 to 69. If we include this population aged 16 to 24 aged 16 to 64 for a given population in the popula- was 69.1 units per 1000 survey year. The figure also tion of potential donors, the persons in 2013, which is includes the collection rate rate of collection drops to much lower than the same for allogeneic WB/RBC approximately 61.2 per cohort’s rate in 2011 (80.8 units collected per 1000 1000 persons aged 16 to per 1000 persons) but total population (all ages) 69. higher than the eligible since 1997. Population esti- population overall (65.5 mates were obtained from units per 1000 persons) and the US Bureau of the Cen- As there was only a 0.8% the rate for persons aged 25 sus.* increase in this US popula- tion cohort between 2011 to 64 (64.7 units per 1000 and 2013, the decline in persons. For persons aged Allogeneic blood collec- the collection rate should 65 or older, the donation tion in the United States be attributed primarily to rate was a much lower 26.7 among AABB member units per 1000 persons.

†Allogeneic blood collection per *http://www.census.gov/popest/ 1000 total population in 2013 data/index.html. was 42.8 units per 1000.

Current Issues in Blood Transfusion 47 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

Figure 8-1. Trends in estimated rates of WB/RBC collection United States, 1980-2013.

US Trends in Utilization decline of 8.5% when com- Blood Availability pared to the member rate of The rate of transfusion was approximately 21.1 units Figure 8-3 illustrates the calculated from the per 1000 in 2011. Extrapo- relationship between allo- national estimate of alloge- lation of the US AABB geneic WB/RBC collec- neic WB/RBC units trans- membership data to the tions and transfusions from fused per 1000 total number of transfusions 1989 to 2013, as well as population of all ages (Fig- throughout the United the margin between units ure 8-2). For this report, a States* yields a rate of collected and those trans- new trend line has been ini- approximately 40.3 units fused. tiated to show the rate of per 1000 population (com- transfusion for AABB US pared to 44.0 units per 1000 in 2011). The trend of increasing members beginning with numbers of collections 2011. reported since 1997 was reversed between 2008 and The US AABB member WB/ 2011, and the decline con- RBC transfusion rate in tinued in 2013, with only *Assuming that the proportion of 2013 was 19.3 allogeneic AABB member contribution is 13.6 million allogeneic RBC units transfused per equivalent in 2013 to that of RBC units collected by 1000 overall population, a 2011. AABB member facilities, for

48 Current Issues in Blood Transfusion THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

Figure 8-2. Trends in estimated rates of WB/RBC transfusion in the United States, 1980-2013.

Figure 8-3. US allogeneic WB/RBC collections and transfusions, 1989-2013.

Current Issues in Blood Transfusion 49 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

a 12.1% decrease from pital population (approxi- (19.9%; general, orthope- 2011. Allogeneic collec- mately 12.8 million units), dic, cardiac surgery, and tions in 2013 were slightly there appears to be only a any other surgery reported higher than those reported very small margin between but not included in these in 1999. Reported amounts collection and transfusion three categories com- transfused in 2013 are only of WB/RBCs. bined), hematology/oncol- slightly higher than those ogy, including stem cell reported in 1999. A total of 34 hospitals transplant (19.2%), and the (6.7%) reported that elec- ICU, including both medi- The available supply of tive surgery was postponed cal and surgical (12.5%). both WB/RBCs and non- due to RBC inventory short- While the actual percent- RBC components was just ages in 2013. Non-surgical ages changed somewhat sufficient to meet overall RBC needs were unmet in from those reported in transfusion demands in 3.4% of reporting hospitals 2011, the largest RBC users 2013. The margin between and special order needs remained general medicine allogeneic WB/RBC supply were unmet in 6.1% of (29.1% in 2011) and sur- and demand depicted in reporting hospitals. When gery (18.2% in 2011). Hos- Figure 8-3 suggests a cor- asked about platelet inven- pitals able to report these rection to the oversupply of tory shortages in 2013, 54 data by hospital service WB/RBC components hospitals (10.5%) reported accounted for only 17.5% observed in previous years that elective surgery had of the US AABB member and a clear movement been postponed. Of WB/RBC transfusions. toward more restricted responding hospitals, blood availability. 22.7% (115 hospitals) The services reporting the reported that they were greatest use of platelet Allogeneic collections in unable to meet other non- products were hematology/ the United States peaked in surgical blood requests. oncology (43.6%), surgery 2008, with 17 million WB/ Non-surgical platelet needs (15.9%; general, orthope- RBC units collected, with a were unmet in 13.2% of dic, and cardiac surgery supply of available units (ie, reporting hospitals. combined), the ICU, units that have passed all including both medical and lab tests, were not wasted, surgical (13.0%), and gen- and are available for trans- Blood Use eral medicine (12.8%). Car- fusion) of 16.9 million. The diac surgery comprised margin between collection Hospitals were asked to 54.5% of the overall plate- and transfusion at that time indicate the number of RBC let use in surgery. As with was 2 million units or and platelet units distrib- RBCs, the largest platelet 12.1% of available collec- uted to individual hospital users in 2013 were the tions. In 2013, there was a services (eg, surgery, hema- same as in 2011: hematol- decrease of 10.3% in avail- tology/oncology, transplant, ogy/oncology (34.4% in able allogeneic collections and ICU) in 2013 (Figures 2011) and surgery (17.6% to 12.9 million units. With 8-4 and 8-5). The services in 2011). an extrapolation of the responsible for the highest AABB number of units use of RBCs were general Of the 219 hospitals that transfused to the total hos- medicine (28.5%), surgery reported blood use by hos-

50 Current Issues in Blood Transfusion THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

Surgery-General 8.6%

Orthopedic Surgery 3.2%

General Medicine 28.5% Cardiac Surgery 7.0%

All Other Surgery Departments 1.0% Transplant (Solid Organ) 1.6%

Trauma/ER 9.5%

ICU (Med+Surg) 12.5%

Hem/Onc/BMT Nephrology/Dialysis 19.2% 2.3% Pediatrics/Neonatology 4.3% OB/Gyn 2.2% Figure 8-4. RBC use by hospital service in 2013.

Surgery-General 5.1% Orthopedic Surgery General Medicine 0.7% 12.8%

Cardiac Surgery All Other Surgery Departments 8.7% 1.5%

Transplant (Solid Organ) 2.4% ICU (Med+Surg) 13.0% Trauma/ER 3.4%

Nephrology/Dialysis 0.3%

Pediatrics/Neonatology 7.9%

OB/Gyn 0.5%

Hem/Onc/BMT 43.6% illbhilii Figure 8-5. Platelet use by hospital service in 2013.

Current Issues in Blood Transfusion 51 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT pital service or department, To calculate the cross- Platelet Age 53.0% determined this by match-to-transfusion ratio location within the hospital. (C:T), the total number of Overall, 286 hospitals The other 47.0% used the allogeneic WB/RBC units (51.8%) responded with an physician service or depart- transfused (6,118,000) was age for apheresis platelets at ment. used as the denominator. the time of transfusion. The overall C:T ratio was Overall, the average age 1.4 crossmatch procedures was 3.2 days at transfusion. per unit transfused, a ratio Crossmatch Procedures The mean reported age was equal to the C:T ratio 3.4 days for the calculated reported in 2011. AABB member hospitals average and 3.1 for the esti- and transfusion services mated average. reported the total number of crossmatch procedures. Red Cell Age Weighted hospital data on Hemovigilance crossmatch procedures The 2013 survey queried indicate that 8,331,000 hospitals to determine the An estimated total of procedures were per- average age of a unit of 27,000 transfusion-related formed in 2013, a decline RBCs at the time of transfu- adverse reactions occurred of 17.4% from the number sion. In this survey, 253 in US AABB member facili- reported by AABB member hospitals responded, repre- ties in 2013, the same num- facilities in 2011 senting approximately ber reported by AABB (10,079,000 procedures) 45.8% of reporting hospi- member hospitals in 2011. and a reflection of the over- tals. The overall mean age Reactions are defined as an all reduction in blood trans- for a red cell unit at transfu- undesirable response or fusion. sion was 19.9 days, compa- effect in a patient that is rable to 2011 (19.2 days). temporally associated with Hospitals were asked to There were 3,717,000 elec- the administration of blood indicate whether they tronic crossmatch proce- or blood component(s) and reported a calculated or dures reported, or 44.6% of that may or may not be the estimated age. The esti- the total procedures result of an incident or an mated mean age was com- reported, a 29.9% increase interaction between a paratively younger at 17.8 over the number of elec- recipient and the blood days than reported in 2011 tronic crossmatch proce- product. The rate of adverse (18.3 days). The calculated dures reported in 2011. reactions is 0.25%, or 2.5 mean age at transfusion was Manual serologic proce- per 1000 units transfused, 22.7 days, older than dures accounted for 36.7% the same rate reported by reported in 2011 (19.4 of the crossmatch proce- AABB member hospitals in days). Only 24.5% (96 hos- dures. Only 2.3% were 2011. This rate is below the pitals) of AABB hospitals reported to be automated range of other national responding to this question serologic crossmatch proce- hemovigilance reporting in 2013 were able to calcu- dures. The remaining pro- systems (3 to 7 events per late the RBC age at transfu- cedures were not 1000 units transfused). It is sion. categorized. almost certain that the explanation for the lower

52 Current Issues in Blood Transfusion THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

US rate is that many that their clinical teams These included specimens adverse reactions are not review reported reactions submitted for ABO testing, reported to the transfusion against the written defini- type and screen, type and service at all. Additional tions (72.6%). crossmatch, prenatal sam- education is needed at all ples, and cord blood sam- levels of the transfusion The rates reported for the ples. Approximately 80% of chain regarding adverse various types of transfusion- AABB member hospitals transfusion reactions and related adverse reactions reported having a policy to the ways to report them for are included in Table 8-1. perform an ABO recheck full implementation of The most commonly before issuing ABO-spe- hemovigilance in the reported transfusion reac- cific blood products (eg, a United States. tions were febrile, non- second patient sample from hemolytic transfusion reac- a separate sample collec- Participants reported tions (1:1244 transfusions) tion). Most hospitals whether they had an elec- and mild to moderate aller- (89.3%) collect data on tronic system for tracking gic reactions (1:1462 trans- sample collection errors, events, which were defined fusions) followed by the including mislabeled sam- as unplanned, unexpected, non-specific “other” ples and wrong blood in and undesired occur- reported transfusion reac- tube errors. Hospitals rences. Nearly 58% of tions (1:1644 transfusions). reported 105,231 sample AABB member hospitals There were significantly collection errors. Of these reported having such a sys- more severe allergic reac- errors, 3339 were wrong tem to track events, which tions reported to US AABB blood in tube (WBIT) errors, is an increase from 48.0% member hospitals in 2013 accounting for 3.2% of reported in 2011. Of these, than in 2011, as well as sample collection errors; most hospitals reported sys- more post-transfusion viral this rate was only 0.03% of tems to track both adverse infections and sepsis all samples processed by reactions and incidents/mis- reported. There were signif- the laboratory, the same takes (39.9%), while 10.0% icantly fewer reports of proportion reported in of hospitals reported having transfusion-associated dys- 2011. Higher rates of a system to track adverse pnea (TAD), delayed hemo- WBITs per sample collected incidents/mistakes only, lytic transfusion reactions, (0.05%), as well as rates of and 8.1% reported having a febrile, hypotensive transfu- WBITs out of all sample system to track adverse sion reactions, delayed collection errors (11.2%) reactions only. Overwhelm- serologic transfusion reac- were seen in the smallest ingly, facilities had written tions, and post-transfusion hospitals (those with surgi- definitions for adverse purpura (Figure 8-6). cal volumes of ≤99 surger- transfusion reactions ies per year). The overall error rate per specimen sub- (96.7%). However, only AABB member hospitals mitted was 1.04%, the 33.7% used the CDC reported 10,130,000 sam- same as reported by AABB National Healthcare Safety ples received in the blood member hospitals in 2011. Network, Hemovigilance bank in 2013, 4.8% more Protocol definitions. Most than reported in 2011. AABB facilities reported

Current Issues in Blood Transfusion 53 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

Table 8-1. Transfusion-Related Adverse Reactions Reported to the Transfusion Service

2011 Reactions: 2013 Reactions: Components Components Transfused Transfused Number of (n=10,962,000 total (n=10,846,000 total Adverse Transfusion Reactions Occurrences 2013 components) components)

Total number of reactions that were 27,077 1:405 1:396 reported to the transfusion service Febrile, nonhemolytic transfusion 8815 1:1244 1:1039 reactions Mild to moderate allergic reactions 7499 1:1462 1:1418 Delayed serologic transfusion reac- 1406 1:7797 1:6557 tions Transfusion-associated circulatory 834 1:13,144 1:12,054 overload (TACO) Delayed hemolytic transfusion reac- 376 1:29,154 1:16,207 tions Hypotensive transfusion reactions 371 1:29,547 1:18,870 Severe allergic reactions 323 1:33,938 1:46,173 Transfusion-associated dyspnea 258 1:42,488 1:26,730 (TAD) Transfusion-related acute lung injury 175 1:62,640 1:55,312 (TRALI) Post transfusion virus transmissions 149 1:73,570 1:658,287 Acute hemolytic transfusion reactions  86 1:127,465 1:99,322 (non-ABO antibodies) Post transfusion sepsis 85 1:128,965 1:276,962 Acute hemolytic transfusion reactions 22 1:498,273 1:412,753 (ABO) Post Transfusion Purpura 9 1:1,218,000 1:132,843 Transfusion-associated graft-vs- 2 1:5,481,000 -- host disease Other reported transfusion reactions 6667 1:1644 N/A

54 Current Issues in Blood Transfusion THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT * Reactions Severe Allergic Mild to Allergic 2013 Reactions Moderate * Virus 2011 Transfusion Transmissions * Febrile Hypotensive PTP Sepsis Post Delayed Serologic Delayed Hemolytic Acute (non-ABO Hemolytic antibodies) (ABO) Hemolytic ons by type, 2011 and 2013. reactions in 2013 TRALI TACO TAD Acute * significantly more * significantly 0 Figure 8-6. Adverse reacti 8000 6000 4000 2000 12000 10000

Current Issues in Blood Transfusion 55

THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

9. Component Costs

For the 2013 AABB Blood All calculations are based the 2011 reported AABB Survey, AABB member hos- on weighted estimates. member average of pitals reported the average Component costs are $225.21. The 2011 NBCUS whole dollar amount their weighted in two parts. First, reported (all hospital) aver- institution paid per unit for each component cost is age estimated cost for a unit the following components: weighted according to the of LR RBCs was $225.42. (1) plasma single donor fro- number of units transfused There were no significant zen within 8 hours of phle- by each facility. As a result, differences in LR RBC costs botomy (FFP), (2) plasma facilities that transfuse by hospital size, although frozen between 8 and 24 larger volumes of a specific hospitals with higher surgi- hours of phlebotomy component will contribute cal volumes tended to pay (PF24), (3) red blood cells more toward the estimated less per unit than those with leukoreduced (LR RBC), average component cost for lower surgical volumes and (4) apheresis platelets that component than will (Table 9-2). leukoreduced (LR). Table 9- facilities with smaller trans- 1 displays a statistical com- fusion volumes. Second, parison of the mean AABB the sampling weights are Plasma hospital cost for each com- applied when calculating ponent in 2013 and 2011. the average, resulting in the The 2013 average dollar The 2011 NBCUS esti- final weighted estimates. amount paid for plasma fro- mated average costs (AABB Year-to-year comparisons zen within 8 hours of phle- member and non-member are made between AABB botomy (FFP) reported by combined) are noted in the member hospitals reporting AABB member hospitals text. In Table 9-2, the aver- in 2011 and AABB mem- was $54.26 per component age AABB hospital compo- ber hospitals reporting in unit (Table 9-1). This nent cost by hospital 2013. decrease of 20.8% com- surgical volume is com- pared to the 2011 average pared to the overall AABB $68.55 was statistically sig- member average. Table 9-3 RBCs nificant. The 2013 AABB displays the Centers for average was 6.3% lower Medicare and Medicaid In 2013, the mean of the than that reported for all Services (CMS) hospital out- average amounts paid hospitals in 2011 ($57.91). patient prospective pay- reported by AABB member There were no significant ment system rates for the hospitals for a unit of LR differences in costs by hos- components. RBCs was $218.87 (Table pital size, although hospi- 9-1) a statistically signifi- tals with higher surgical cant decrease of 2.8% from volumes tended to pay less

Component Costs 57 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

Table 9-1. Mean AABB Member Hospital Amount ($) Paid per Selected Component Unit in 2011-2013*

% Change Component AABB 2013* AABB 2011* 2011-2013 p-value Red Blood Cells, leukoreduced $218.87 225.21† –2.8 0.005 Plasma, frozen within 8 hours of  $ 54.26 68.55‡ –20.8 <0.001 phlebotomy (FFP) Plasma, frozen between 8 and 24 hours  $ 53.03 59.75§ –11.2 <0.001 of phlebotomy (PF24) Apheresis platelets, leukoreduced $516.96 571.62|| –9.6 <0.001

*Calculations are based on weighted estimates, which generally differ from the unweighted estimates by less than $1.00. †2011 All hospitals LR RBC average cost $225.42. ‡2011 All hospitals FFP average cost $57.91. §2011 All hospitals PF24 average cost $56.08. ||2011 All hospitals apheresis platelets average cost $535.17.

per unit than those with Apheresis Platelets Reimbursement lower surgical volumes (Table 9-2). For a unit of apheresis The 2013 CMS hospital platelets, in 2013 AABB outpatient prospective pay- The hospital cost for plasma member hospitals paid an ment system (OPPS)* reim- frozen between 8 and 24 average of $516.96, which bursement rates for the hours (PF24) after phlebot- was a statistically signifi- blood components assessed omy averaged $53.03 for cant reduction of 9.6% are reported in Table 9-3. AABB member hospitals compared to the $571.62 Changes to hospital costs (Table 9-1). This is signifi- cost of a unit reported by for the different compo- cantly lower (11.2%) than AABB member hospitals in nents ranged from the 2011 AABB member 2011 (Table 9-1). The 2011 decreases of 2.8% to 20.8% hospital average of $59.75. NBCUS reported average between 2011 and 2013 The average cost reported cost for a unit of apheresis (Table 9-1). The CMS OPPS for all hospitals in 2011 was platelets was $535.17. reimbursement rate adjust- $56.08 (5.3% greater than Again, there were no signif- ments ranged from that reported by AABB hos- icant differences in hospi- decreases of 0.8% to 5.1%. pitals in 2013). There were tals costs reported by AABB no significant differences in member hospitals of differ- costs by hospital size for ent surgical volume (Table plasma frozen between 8 9-2). *Department of Health and and 24 hours. Human Services. Medicare Pro- gram; Changes to the Hospital Outpatient Prospective Payment System and Calendar Year 2013 Payment Rates; Final rule with comment period.

58 Component Costs THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT is the maxi- Apheresis Platelets (LR) Platelets umber reported here umber reported 24 hours 24 between 8 and between Plasma, frozen frozen Plasma, within 8 hours Plasma, frozen pitals did n to all not The questions. answers provide Avg ($) Avg ($) Avg ($) ($) Avg mponent Cost ($) by Surgical Volume, 2013 (LR) RBCs 215.67 to 238.17 52.84 to 65.65to 50.52 64.95 417.38555.56 to nt varies because some hos varies because some nt the six survey questions. the six survey hospital costs 998,000 8 65 229.27 65.11 216.18 52.84 64.95 417.38 50.52 499.88 Table 9-2. Average AABB Member Hospital Co Surgical Annual Volume 100-9991,000-1,3991,400-2,3992,400-4,999Hospitals* No. of 5,000-7,999 UnknownAll Hospitals 30 38Range of average 52 90 61 each number of responses for blood compone *The 238.17 229.93 15 359 226.33 220.37 215.67 65.65 61.05 219.67 218.87 60.10 54.99 54.77 58.46 57.03 59.93 54.26 58.15 55.43 54.46 548.82 555.56 59.58 53.03 542.94 545.17 538.42 523.47 516.96 mum number of responses over

Component Costs 59 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT Rate  yment Rates; Final Final Rates; yment % Difference Reimbursement Between Hospital Average Paid Paid and Average alendar Year 2011 alendar Year 2011 Pa e with comment period. % Change ; HCPCS = Health-care Common Procedure ; HCPCS = Health-care (2013-2011) † ocedural Terminology ocedural Calendar Year 2013 Payment Rates; Final rul Payment Year 2013 Calendar ices; CPT = Current Pr Paid Reimbursement Rate AABB AABB Hospital Average $ $ Average re and Medicaid Serv P9017P9059 9508 0955 54.26 53.0379.35 73.15 78.71 75.53 –0.8 –3.345.1 42.4 Prospective Payment System Rates for Selected Blood Components Reimbursement Code Reimbursement Outpatient Prospective Payment System and and System Payment Prospective Outpatient  Services. Medicare Program; Changes to the Hospital Outpatient Prospective Payment System and C and System Payment Prospective Outpatient Hospital to the Changes Program; Medicare Services.  phlebotomy of phlebotomy Medicare Program; Changes to the Hospital to the Hospital Changes Program; Medicare Blood ComponentBlood CPT/HCPCS APC 2013 2011* 2013 Coding System. Table 9-3. CMS Hospital Outpatient Red Blood Cells (leukoreduced)offrozen withinPlasma 8 hours P9016*Department of Health and Human 0954† 218.91 194.86 193.24 –0.8 –11.7 Plasma frozenand betweenPlasma 24 hours 8 Apheresis platelets (leukoreduced) P9035 9501 516.96 538.51 511.27 –5.1–1.1 APC = Ambulatory Patient Classification; CMS = Centers for Medica for = Centers CMS Classification; Patient = Ambulatory APC rule with comment period. rule

60 Component Costs THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

Table 9-3 shows that a unit able measure of Medicare tal costs for LR RBCs, FFP, of apheresis platelets was reimbursement for individ- PF24, and apheresis plate- reimbursed at approxi- ual blood components. lets were significantly lower mately 98.9% of the aver- Most Medicare reimburse- in 2013 compared to AABB age cost paid by AABB ment for blood is part of the member hospitals in 2011. hospitals. The reimburse- diagnosis-related group ment for a unit of LR RBCs (DRG) payment made for In the US, the most com- in 2013 was approximately inpatient services and is monly transfused blood 88.3% of the average hospi- impossible to tease apart product is RBCs, a large tal cost. A unit of FFP was from the other aspects of proportion of which are  reimbursed at 45.1% more the DRG. Other payers LR. Hospitals are under- than the reported average besides Medicare pay for reimbursed by 11.7% per amount paid by AABB blood by using various unit for the acquisition cost member hospitals in 2013. mechanisms that are not alone of this product, not PF24 was reimbursed at included in this report. including other costs asso- 42.4% more than the ciated with its transfusion. reported average cost to In 2013, AABB members AABB member hospitals in Summary hospitals transfused over 2013. 6.1 million RBCs, this alone The average hospital costs represents an underpay- CMS OPPS rates are for blood components were ment to hospitals of $157 reported here because they lower in 2013 compared to million. are the only clearly identifi- 2011. Average AABB hospi-

Component Costs 61

THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

10. Component Modification

Leukocyte Reduction ponents due to the overall AABB Member decline in collections. Transfusion of Irradiated There was a drop in the per- Blood components are leu- and LR Components centage of WB/RBC units kocyte reduced (LR) to undergoing LR (78.5%) decrease the risk of febrile compared to 2011, when Table 10-3 summarizes the non-hemolytic reactions, 85.2% of the WB/RBC units types and number of irradi- transmission of cytomegalo- were LR. More WBD plate- ated and LR blood compo- virus (CMV) infection, and lets were reported to be LR nent units transfused during HLA alloimmunization that in 2013 both in number 2013. Blood centers and may lead to platelet refrac- (638,000) and in percent- AABB member hospital toriness. Leukocyte reduc- age of those distributed transfusion services tion can be performed (77.9% vs. 57.2% in 2011). reported transfusion of a during collection, at some Nearly all apheresis plate- total of 1,796,000 irradi- time before components are lets were LR (92.7%). There ated units in 2013, repre- placed into inventory, and were few reports of other senting 32.9% of all after storage; these are cate- LR components in the 2013 components transfused by gorized as “before or after survey data. facilities reporting irradi- storage, but not at bedside” ated units. leukocyte reduction. A total of 12,804,000 blood com- In comparison with 2011, ponents (80.4% of all RBCs blood centers produced A total of 5,486,000 LR and platelets) were LR 11.2% fewer LR compo- components were trans- before storage in 2013 by nents than in 2011; this is fused by blood center and AABB member blood cen- consistent with the 12.1% hospital transfusion ser- ters and hospitals that col- decline in WB/RBC collec- vices. Of all LR units trans- lect blood (Table 10-1). tion overall. AABB member fused, 99.5% were LR Components can also be LR hospitals produced 29.4% before or after storage (not bedside during the time of fewer LR components in at bedside), and only 0.5% transfusion, though this 2013 when compared with were leukocyte filtered at practice is waning (Figure 2011 (Table 10-2). Overall, bedside. In 2013, substan- 10-1). the number of LR compo- tial proportions of all trans- nents prepared decreased fused RBCs and platelets by 12.1% from 2011, how- were reported to have been It is important to focus on ever, the percentage of WB/ LR. Among the hospitals the percentage of LR com- RBC and platelet compo- reporting both numbers of ponents rather than the nents that were LR units transfused and those actual number of LR com- decreased by only 2.7%. transfused which had been

Component Modification 63 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT Components Leukocyte-Reduced % of Total Available Total Available Available Components Reduced Reduced Leukocyte- Prestorage .5 12,229,000 14,352,000 85.2 Components Leukocyte-Reduced % of Total Available torage Leukocyte Reduction in US AABB Facilities ,000 92.7 1,967,0002,283,000 86.2 2013 2011 Components Total Available Total Available 12,804,00012,804,000 20,741,000 15,916,000 61.7 80.4 14,571,000 22,642,000 14,564,000 17,278,000 64.4 84.3 Prestorage Leukocyte-Reduced Blood Component WB-Derived PlateletsWB-Derived PlateletsApheresis UnitsComponent Other Total Components 638,000 TotalWB/RBCs + Platelets <1,000 2,063,000819,000 4,825,000 2,226 77.9 0.0 368,000643,000 7,000 57.2 5,364,000 0.1 Table 10-1. Blood Components Modified to Achieve Pres WB/RBCs 10,103,000 12,871,000 78

64 Component Modification THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT Total RBCs Modified by LR by RBCs Modified Total Transfused RBCs at Bedside) (not Leukoreduced Transfused RBC Units (at bedside) Leukofiltered Transfused RBC Units Irradiated 1999 2001 2004 2006 2008 2011 AABB 2011 AABB 2013 0

8,000 6,000 4,000 2,000

16,000 14,000 12,000 10,000 Units in Thousands Thousands in Units Figure 10-1. US Transfusion of modified RBC components.

Component Modification 65 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT 000 –29.4% 12,804,000 14,571,000 –12.1% Achieve Pre-Storage Leukocyte Reduction by Facility Type Blood CentersBlood Hospitals AABB Facilities AABB All 2013 2011 % Change 2013 2011 % Change 2013 2011 % Change 12,286,000 13,837,000 –11.2% 518,000 734, cyte reduced before storage (not at the bedside) from 2011 to 2013 Table 10-2. Change in Number of Blood Components Modified to Components leuko-

66 Component Modification THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT Leukocyte Total Units Transfused Reduced: % of Transfused by US Transfused Irradiated: % % Irradiated: of Total Units Total reported Transfused facilites that components) Leukoreduced Leukoreduced (only includes Leukocyte Reduction and Total reported Irradiated Irradiated Transfused Transfused facilites that facilites components) (only includes includes (only Total Total Units Reduced 5,486,000 5,466,000 6,120,000 32.9 89.6 847,000 837,000 884,000 60.7 95.8 Leukocyte Leukocyte Filtration at Filtration Components (Reduced by the Bedside) mponent Units Modified by Irradiation or /leukoreduced units as well as total well total units as components. as /leukoreduced Storage Reduced Bedside) Leukocyte Leukocyte (not at the Components Components Before or After 35,000 37,000 2,000 8,000 <1,000 37,000 <1,000 57,000 8,000 58,000 3,000 34,000 61.4 66.7 63.8 23.5 332,000 389,000 102,000 8,000122,000 397,000 422,000 1,000 440,000 123,000 175,000 78.7 173,000 90.2 58.3 71.1 Irradiated transfusion of irradiated transfusion Components  Aliquots Aliquots (individual units) Aliquots Pediatric Plasma Units or Pediatric Total Components 1,796,000 5,460,000 26,000 Apheresis Platelets 508,000 Platelet Units or Pediatric 846,000 <1,000 Table 10-3. Estimated Number of Blood Co AABB Member Facilities in 2013* Component Blood WB/RBCs Platelets WB-Derived RBC Units or Pediatric 817,000 4,058,000 16,000 4,074,000 *Includes facilities that reported 3,972,000 4,531,000 20.6 89.9

Component Modification 67 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

LR, 89.9% of WB/RBCs, while the number of irradi- Transfusing facilities were 90.2% of WBD platelets, ated units transfused asked whether they had a and 95.8% of apheresis decreased by 16.8%, the policy in 2013 to transfuse platelets were LR. While the percentage of all RBC units only LR components: percentage of transfused transfused that were irradi- 71.8% responded they did apheresis platelets appears ated (13.3%) decreased by have such an LR-only pol- discrepant to the apheresis 1.5% from 2011 (14.8%) icy in place; this was platelet collection process among AABB member hos- unchanged from 2011. and the numbers of LR pitals. Both the total num- Another 11.7% of hospitals apheresis platelets reported ber and percentage of all had a policy to transfuse produced, transfusion ser- units transfused that were only LR components to par- vices, or those completing LR declined in 2013. There ticular types of patients. the survey, may not have were 13.7% fewer LR RBC Among institutions with been aware of the LR pro- units transfused and the these patient-specific LR cessing that occurs during proportion of RBC units policies, 84.6% reported LR collection. transfused as LR declined policies for oncology by 4.1%. Reports of bedside patients, 81.5% for neo- Table 10-4 and Figure 10-1 LR of RBCs declined to only nates, 47.7% for cardiovas- summarize the trends in 16,000 total transfusions, a cular surgery patients irradiated and LR RBC com- 62.8% decrease from 2011. (compared to 13.8% in ponents transfused. 2011), and 61.5% for other Between 2011 and 2013, patients.

Table 10-4. Total Number of Irradiated and Leukocyte Reduced Red Blood Cell (RBC) Units Transfused by US AABB Member Hospitals in 2013, compared with RBC Units Transfused by AABB Member Hospitals in 2011

Red Blood Cell Units

Change 2013 - Modification 2013 2011 2011 % Change

Irradiated 817,000 982,000 –165,000 –16.8 Leukocyte-Reduced, total 4,074,000 4,719,000 –645,000 –13.7 Before or After Storage (not at the bedside) 4,058,000 4,676,000 –618,000 –13.2 At the bedside 16,000 43,000 –27,000 –62.8

68 Component Modification THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

11. Appendix I: Methods

Survey Population US Military Blood tals were stratified into eight Program Response categories based on the 2013 annual inpatient sur- The 2013 AABB Blood Sur- gical volume: 99 surger- vey was distributed through Response rates for the offi- ies; 100-999 surgeries; a web-based survey tool* cial US military blood pro- 1000-1399 surgeries; 1400- from November 7, 2014, to gram or the Armed Services 2399 surgeries; 2400-4999 February 16, 2015, to all blood programs are pre- surgeries; 5000-7999 sur- AABB institutional mem- sented in Table A-2. Data geries; ≥8000 surgeries; bers. The two broad catego- from these facilities were and those having unknown ries of participants were: analyzed separately from surgical volume. the US blood center • Blood Centers analysis. Blood Centers were strati- • Hospitals fied into three categories AABB US Member Data based on the number of Non-US AABB Member WB/RBCs collected in Analysis 2013: >250,000 units; Response 50,000 to 249,999 units; The US hospital population and <50,000 units.  Response rates for the non- included hospitals located Table A-3 summarizes the US facilities are presented within the 50 states and the survey population and in Table A-1. Because data District of Columbia.† The response rates. received from these mem- report includes weighted bers were sparse, data from aggregate reported results. these facilities were neither Response Rates and imputed nor weighted. In For the purpose of stratifica- Weights future surveys, AABB will tion, the AABB member approach non-US mem- hospitals were matched A responding institution bers directly to ensure it is with the American Hospital understood that all mem- could provide data for itself Association (AHA) hospital and also for other institu- bers are invited to partici- membership information pate. tions. This resulted in inclu- database to record the num- sion of some institutions ber of inpatient surgeries that were not in the original performed in 2013. Hospi- survey population (eg, not current AABB members). †Includes hospitals from the US Information collected about *http://www.qualtrics.com. Department of Veterans Affairs.

Appendix I: Methods 69 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

Table A-1. AABB Blood Survey 2013: Survey Population and Response Rates—Non-US Members

Institution Type Respondents Total Response Rate (%) Non-US Blood Centers 6 17 35.3 Non-US Hospitals 8 35 22.9

Table A-2. AABB Blood Survey 2013: Survey Population and Response Rates—US Military Blood Program

Institution Type Respondents Total Response Rate (%) US Military Blood Program 2 4 50.0

Table A-3. AABB Blood Survey 2013: Survey Population and Response Rates

Institution Type Respondents Total Response Rate (%) Blood Centers WB/RBC Collections 250,000+ 7 7 100.0 WB/RBC Collections 50,000 to 249,999 33 36 91.7 WB/RBC Collections <50,000 33 36 91.7 Hospitals 99 surgeries per year 48 62 77.4 100 - 999 surgeries per year 73 127 57.5 1000 - 1399 surgeries per year 38 67 56.7 1400 - 2399 surgeries per year 68 141 48.2 2400 - 4999 surgeries per year 122 284 43.0 5000 - 7999 surgeries per year 86 178 48.3 8000 surgeries per year 84 152 55.3 Unknown surgical volume 33 57 57.9 Total Facilities 625 1147 54.5

these institutions was not as summarized in Table A-4 more than 50.0% of the complete as it would have and discussed in the section decrease in the total popu- been if these hospitals had below. In the 2013 survey, lation of blood centers from completed an individual a blood center that had pre- 136 to 79. survey. Overall response viously reported by regions rates for 2013 and for previ- or sub-units, reported in the ous blood survey years are aggregate. This represents

70 Appendix I: Methods THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

Table A-4. Response Rate Summary and Comparison with Previous Surveys

2011 AABB Facility Type 2013 Cohort 2011† 2009† 2007† Blood Centers 92.4% 96.3% 96.3% 93.3% 91.4% (73/79) (131/136) (131/136) (126/135) (128/140) Hospitals (all) 51.7% 49.5% 42.3% 51.5% 59.9% (552/1068) (581/1173) (1342/3175) (1529/2970) (1707/2848) Total 54.5% 54.4% 44.1% 53.1% 61.3% (625/1147) (712/1309) (1493/3381)* (1660/3129)* (1849/3015)*

*Includes Cord Blood Banks. †NBCUS Total population.

Blood Centers tions (estimated from 2011 pitals were reported by survey data or obtained blood centers and data for After excluding blood cen- separately for the non- 15 hospitals were submitted ters that ceased operations, respondents) as shown in by other responding hospi- were unable to provide Table A-5. tals. For the purpose of complete data, merged with weighting, these 15 hospi- other blood centers, or tals were treated as if they were affiliated with the US Hospitals were in the original survey Armed Force services, the population. final survey population It was difficult to weight hospital data because some included 79 blood centers. Table A-6 summarizes the data were reported by other The weights were calcu- weights for 2013 hospitals hospitals. In addition, blood lated as the inverse of the by inpatient surgical vol- centers could also report probability of response. The ume. The weight is the transfusion information for weights were adjusted for inverse of the probability of hospitals when they served nonresponse by stratifying response. the blood centers by the as centralized transfusion number of WB/RBC collec- services. Data for 146 hos-

Table A-5. Weights for Blood Centers

Volume of RBC Collections Total Respondents Weight WB/RBC Collections 250,000+ 7 7 1.0000 WB/RBC Collections 50,000 to 249,999 36 33 1.0909 WB/RBC Collections <50,000 36 33 1.0909 Total 79 74

Appendix I: Methods 71 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

Table A-6. Weights for 2013 Hospitals

Surgical Volume Non-Respondents Respondents Total Weight 99 surgeries per year 14 48 62 1.2917 100 - 999 surgeries per year 54 73 127 1.7397 1000 - 1399 surgeries per year 29 38 67 1.7632 1400 - 2399 surgeries per year 73 68 141 2.0735 2400 - 4999 surgeries per year 162 122 284 2.3279 5000 - 7999 surgeries per year 92 86 178 2.0698 8000 surgeries per year 68 84 152 1.8095 Unknown surgical volume 24 33 57 1.7273 Total 516 552 1068

Imputation of Data Items they would have been if an ables for the 2013 survey individual survey had been responses. completed separately. Data Apportionment Characterization of Item Imputation A responding institution Respondents could provide data for itself and also for other institu- Facilities were contacted to tions. A new record was verify if the submitted data Among the responding created for each institution reflected their current blood blood centers, 18 blood for which information collection and transfusion centers (24.3%) self-defined could be obtained and veri- model/program, to avoid as centralized transfusion fied as being a population inappropriate imputation. services, which was a of interest (reported by an This restricted item imputa- numeric decrease from the AABB member but not nec- tion to only two critical 21 blood centers reporting essarily an active AABB questions on responding this characteristic in 2011, institutional member). For blood center and hospital but was an increased per- each new record created, surveys: total numbers of centage of centers conduct- certain categorical data ele- allogeneic units discarded ing this activity (18.3% in ments from the responding in 2013 for reactive infec- 2011). Among the 21 blood institution’s response were tious disease markers and centers with centralized copied (eg, “Were any elec- for all other reasons. Miss- transfusion service activity tive surgeries postponed ing values were imputed on in 2011, 12 centers due to RBC inventory short- 12 responding blood cen- reported continuing this ages in 2013 (yes/no)?”) and ters and 19 responding hos- activity in 2013. other data were appor- pitals, for a total of 46 tioned (eg, total number of imputed data items. In all A total of 65 (11.8%) allogeneic RBCs trans- cases, the imputation was responding hospitals fused). In some cases, the based on models of rela- reported collection data data collected in this man- tionships between vari- and characterized them- ner were not as complete as selves as a hospital-based

72 Appendix I: Methods THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT blood bank and transfusion member institutions from For hospitals, the results service that collects blood. the 2011 NBCUS were con- from the 2013 AABB Blood Among 2011 AABB hospi- sidered for analysis. Survey were compared with tals, 94 (16.2%) of the Although some mergers adjusted 2011 NBCUS responding hospitals occurred among blood cen- results, so as to include reported collecting blood ters since 2011, we were AABB member hospitals products. able to collect separate data only. The weights* for 2011 for most of the blood cen- AABB respondent hospitals ters that were active indi- were adjusted by calibrat- Comparison with vidually in 2011. A large ing their final weights by Previous Surveys blood center with multiple the AABB member hospital regions reported 2013 data population for the year in the aggregate rather than 2011 as shown in  The results from 2013 by region as had been done Table A-7. AABB Blood Survey were previously. This suggested compared with the AABB an apparent, but not an member cohort from the actual drop in the number *Weight calibration 2011 NBCUS. All AABB of blood centers reporting. ratio=0.902(1173/1301)

Table A-7. Calibrated Average Final Weight by USPHS Region and Inpatient Surgical Volume – 2011 AABB Hospital Respondents

By USPHS Region Region Average Original Weight Average Calibrated Weight I 2.5561 2.3056 II 2.1217 1.9138 III 2.2184 2.0010 IV 3.1917 2.8789 V 3.2788 2.9575 VI 3.7305 3.3649 VII 3.0672 2.7666 VIII 3.5883 3.2366 IX 3.5486 3.2008 X 3.0833 2.7811 By Inpatient Surgical Volume Surgical Volume Average Original Weight Average Calibrated Weight 100 - 999 6.7204 6.0618 1,000 - 1,399 2.3875 2.1535 1,400 - 2,399 2.5096 2.2637 2,400 - 4,999 2.2815 2.0579 5,000 - 7,999 1.9086 1.7216 8,000 2.015 1.8175 Unknown 2.213 1.9961

Appendix I: Methods 73 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

Data Quality Control Statistical Analysis The timing of the 2013 AABB Blood Survey release Approximately 11 data ele- The final weighted esti- (November 2014) com- ments from survey Section mates for the 2013 AABB peted with the release of B (blood collection, pro- Blood Survey were com- the 2013 NBCUS released cessing, and testing), and pared with the weighted through the Centers for Dis- four data elements from sur- estimates from the AABB ease Control and Preven- vey Section C (blood trans- cohort of 2011 NBCUS. tion (CDC) in December fusion) were analyzed Continuous variables were 2014. This was a major lim- rigorously for data quality. compared using student-t iting factor for data collec- Any facility that reported tests. Categorical variables tion. In addition, the 2013 values beyond the expected were compared using chi- AABB Blood Survey was threshold was contacted  square tests of proportions conducted only among via email and telephone to with p<0.05. Statistical AABB institutional mem- confirm data accuracy. analyses were carried out bers; therefore, the esti- Thirty-four blood centers using SAS 9.4. mates from this survey are and 77 hospitals reported predictive of AABB member corrected values during  transfusion activity in the the data quality control Limitations United States, but not nec- process. essarily of overall US national transfusion activ- The relatively low response ity. The uncertainty due to Outliers for all quantitative rate of hospitals in the 2013 non-sampling errors such as variables were also AABB Blood Survey was a measurement errors (eg, reviewed. Any data which limitation. Non-response data entry errors, misread- were uncharacteristic outli- increases the uncertainty of, ing of questions, or missed ers were investigated with or variability in, the esti- entry) and the variability the reporting facility. mates, and it may also due to item imputation  introduce bias to the esti- are not reflected in the mates, the extent of which estimates. is very difficult to assess.

74 Appendix I: Methods THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

12. Appendix II: Glossary of Terms

Terms Definition Apheresis Donation A procedure where WB is removed from the body and desired component(s), such as plasma or platelets is retained and the remainder of the blood is returned to the donor. Autologous Blood Donation Self-directed donations. Blood collected from an individual with the intention to be given back to the same individual. Centralized transfusion service A hospital or blood center that collects blood from donors and supplies blood, com- ponents, medical services and/or cross matched blood products to multiple trans- fusing facilities. Collected Successful whole blood or apheresis collections placed into production (not QNS, or other removals). Community In this survey refers to those allogeneic donations not directed to a specific patient. Deferrals The number of individuals who present to donate blood or blood components and are unable to donate for specific reasons (eg, low hemoglobin, high risk behavior). Directed Donation Allogeneic donations intended for a specific patient. Donation The collection of a unit of blood or blood component from a volunteer donor. First-time allogeneic donor A donor who is donating for the first time at your center. INR/PTT INR (international normalized ratio)/PTT (Partial thromboplastin time). INR is a sys- tem established by the World Health Organization (WHO) and the International Committee on Thrombosis and Hemostasis for standardizing the results of pro- thrombin time tests. PTT is a blood test that measures the time for a patient’s blood to clot. Generally, a PTT test is used to detect and diagnose a bleeding disorder or excessive clotting disorder, while INR is calculated from a PTT result to monitor the effectiveness to anticoagulants. Modify Used in this survey to refer to procedures applied by a blood center, hospital blood bank, or transfusion service that may affect the quality or quantity of the final prod- uct (eg, irradiation, leukofiltration, or production of aliquots of lesser volume). Outdated Units that expire on your shelf. Patient Blood Management An evidence-based, multidisciplinary approach to optimizing the care of patients who might need transfusion. PBM encompasses all aspects of patient evaluation and clinical management surrounding the transfusion decision-making process, including the application of appropriate indications, as well as minimization of blood loss and optimization of patient red cell mass. PCC Prothrombin complex concentrates (PCC) are an inactivated concentrate of Vitamin-K dependent blood clotting factors II, VII, IX and X, as well as protein C  and S, prepared from fresh-frozen human blood plasma, used to reverse the effects of anticoagulants.

Appendix II: Glossary of Terms 75 THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

Performance benchmarking A program designed to compare the performance of an individual hospital on one or programs more metrics with others on a national, regional, or hospital system-wide basis (eg, Q-Probes, Premier).

Plasma A) Plasma frozen within 24 hours After phlebotomy: plasma separated from the blood of an individual donor and placed at –18 C or colder within 24 hours of collection from the donor. Sometimes also referred to as PF24. B) Plasma, Jumbo: for the purposes of this survey, FFP having a volume greater than 400 mL. C) FFP: Fresh Frozen Plasma. Plasma frozen within 8 hours of collection. Present to Donate A person presents to donate when he or she initiates the donation process through appearance and registration at a donation site. Produced Blood component manufactured from a unit of whole blood. Recipient A unique individual patient receiving a transfusion one or more times in a calendar year. Released for Distribution Units that have fulfilled all testing and processing requirements and have been made available to customers for transfusion. Repeat allogeneic donor A donor who has previously donated a blood component for community use, using your facility’s definition. Severe Donor-Related Adverse Severe adverse events occurring in donors attributed to the donation process that Reactions include, for example, major allergic reaction, arterial puncture, loss of conscious- ness of a minute or more, loss of consciousness with injury, nerve irritation, etc. Transfusion Related Adverse An undesirable response or effect in a patient temporally associated with the admin- Reactions istration of blood or blood components. It may or may not be the result of a mis- take. Transfusion Service A facility that performs, or is responsible for the performance of, the storage, selec- tion, and issuance of blood and blood components to intended recipients. Wasted components Include components that were ordered but not used, out of temperature, broken bags, etc. Do NOT include outdated components. Whole Blood (WB) Donation A blood donation or sample of blood taken from the venous or arterial circulation that is composed of blood cells, platelets, and plasma.

76 Appendix II: Glossary of Terms THE 2013 AABB BLOOD COLLECTION, UTILIZATION, AND PATIENT BLOOD MANAGEMENT SURVEY REPORT

13. Acknowledgements

AABB acknowledges the following individuals who contributed their time and expertise to this project:

AABB Staff AABB Volunteers Nina Salamon Dr. Beth Shaz Frank Berry Dr. James P. AuBuchon Dr. Zbigniew Szczepiorkowski Russell Cotten Dr. Richard Benjamin Dr. Lynne Uhl Erika Hastings Greg Bishop Dr. Jonathan Waters Nina Hutchinson Dr. Kelley Capocelli Dr. Naynesh Kamani Dr. Claudia Cohn Dr. Stephen Kleinman Dr. Louis Katz Other Contributors Miriam A. Markowitz, MSc Dr. Naomi Luban Dr. Susan Hinkins, NORC Laurie Munk, MLS Dr. Kathleen Puca

Acknowledgements 77